Merge branch 'BerriAI:main' into ollama-image-handling

.circleci/config.yml

@@ -41,6 +41,7 @@ jobs:
             pip install langchain
             pip install lunary==0.2.5
             pip install "langfuse==2.27.1"
+            pip install "logfire==0.29.0"
             pip install numpydoc
             pip install traceloop-sdk==0.0.69
             pip install openai
@@ -86,7 +87,6 @@ jobs:
               exit 1
             fi
             cd ..
-  
 
       # Run pytest and generate JUnit XML report
       - run:
@@ -94,7 +94,7 @@ jobs:
           command: |
             pwd
             ls
-            python -m pytest -vv litellm/tests/ -x --junitxml=test-results/junit.xml --durations=5 
+            python -m pytest -vv litellm/tests/ -x --junitxml=test-results/junit.xml --durations=5
           no_output_timeout: 120m
 
       # Store test results
@@ -170,6 +170,7 @@ jobs:
             pip install "aioboto3==12.3.0"
             pip install langchain
             pip install "langfuse>=2.0.0"
+            pip install "logfire==0.29.0"
             pip install numpydoc
             pip install prisma            
             pip install fastapi            
@@ -222,7 +223,7 @@ jobs:
           name: Start outputting logs
           command: docker logs -f my-app
           background: true
-      - run: 
+      - run:
           name: Wait for app to be ready
           command: dockerize -wait http://localhost:4000 -timeout 5m
       - run:
@@ -230,7 +231,7 @@ jobs:
           command: |
             pwd
             ls
-            python -m pytest -vv tests/ -x --junitxml=test-results/junit.xml --durations=5 
+            python -m pytest -vv tests/ -x --junitxml=test-results/junit.xml --durations=5
           no_output_timeout: 120m
 
       # Store test results
@@ -252,7 +253,7 @@ jobs:
           name: Copy model_prices_and_context_window File to model_prices_and_context_window_backup
           command: |
             cp model_prices_and_context_window.json litellm/model_prices_and_context_window_backup.json
-      
+
       - run:
           name: Check if litellm dir was updated or if pyproject.toml was modified
           command: |
@@ -337,4 +338,4 @@ workflows:
           filters:
             branches:
               only:
-                - main
+                - main

.git-blame-ignore-revs

@@ -0,0 +1,10 @@
+# Add the commit hash of any commit you want to ignore in `git blame` here.
+# One commit hash per line.
+#
+# The GitHub Blame UI will use this file automatically!
+#
+# Run this command to always ignore formatting commits in `git blame`
+#     git config blame.ignoreRevsFile .git-blame-ignore-revs
+
+# Update pydantic code to fix warnings (GH-3600)
+876840e9957bc7e9f7d6a2b58c4d7c53dad16481

.github/pull_request_template.md

@@ -1,6 +1,3 @@
-<!-- This is just examples. You can remove all items if you want. -->
-<!-- Please remove all comments. -->
-
 ## Title
 
 <!-- e.g. "Implement user authentication feature" -->
@@ -18,7 +15,6 @@
 🐛 Bug Fix
 🧹 Refactoring
 📖 Documentation
-💻 Development Environment
 🚄 Infrastructure
 ✅ Test
 
@@ -26,22 +22,8 @@
 
 <!-- List of changes -->
 
-## Testing
+## [REQUIRED] Testing - Attach a screenshot of any new tests passing locall
+If UI changes, send a screenshot/GIF of working UI fixes
 
 <!-- Test procedure -->
 
-## Notes
-
-<!-- Test results -->
-
-<!-- Points to note for the reviewer, consultation content, concerns -->
-
-## Pre-Submission Checklist (optional but appreciated):
-
-- [ ] I have included relevant documentation updates (stored in /docs/my-website)
-
-## OS Tests (optional but appreciated):
-
-- [ ] Tested on Windows
-- [ ] Tested on MacOS
-- [ ] Tested on Linux

docs/my-website/docs/caching/all_caches.md

@@ -1,7 +1,7 @@
 import Tabs from '@theme/Tabs';
 import TabItem from '@theme/TabItem';
 
-# Caching - In-Memory, Redis, s3,  Redis Semantic Cache
+# Caching - In-Memory, Redis, s3, Redis Semantic Cache, Disk
 
 [**See Code**](https://github.com/BerriAI/litellm/blob/main/litellm/caching.py)
 
@@ -11,7 +11,7 @@ Need to use Caching on LiteLLM Proxy Server? Doc here: [Caching Proxy Server](ht
 
 :::
 
-## Initialize Cache - In Memory, Redis, s3 Bucket, Redis Semantic Cache
+## Initialize Cache - In Memory, Redis, s3 Bucket, Redis Semantic, Disk Cache
 
 
 <Tabs>
@@ -159,7 +159,7 @@ litellm.cache = Cache()
 # Make completion calls
 response1 = completion(
     model="gpt-3.5-turbo", 
-    messages=[{"role": "user", "content": "Tell me a joke."}]
+    messages=[{"role": "user", "content": "Tell me a joke."}],
     caching=True
 )
 response2 = completion(
@@ -174,6 +174,43 @@ response2 = completion(
 
 </TabItem>
 
+<TabItem value="disk" label="disk cache">
+
+### Quick Start
+
+Install diskcache:
+
+```shell
+pip install diskcache
+```
+
+Then you can use the disk cache as follows.
+
+```python
+import litellm
+from litellm import completion
+from litellm.caching import Cache
+litellm.cache = Cache(type="disk")
+
+# Make completion calls
+response1 = completion(
+    model="gpt-3.5-turbo", 
+    messages=[{"role": "user", "content": "Tell me a joke."}],
+    caching=True
+)
+response2 = completion(
+    model="gpt-3.5-turbo", 
+    messages=[{"role": "user", "content": "Tell me a joke."}],
+    caching=True
+)
+
+# response1 == response2, response 1 is cached
+
+```
+
+If you run the code two times, response1 will use the cache from the first run that was stored in a cache file.
+
+</TabItem>
 
 </Tabs>
 
@@ -191,13 +228,13 @@ Advanced Params
 
 ```python
 litellm.enable_cache(
-    type: Optional[Literal["local", "redis"]] = "local",
+    type: Optional[Literal["local", "redis", "s3", "disk"]] = "local",
     host: Optional[str] = None,
     port: Optional[str] = None,
     password: Optional[str] = None,
     supported_call_types: Optional[
-        List[Literal["completion", "acompletion", "embedding", "aembedding"]]
-    ] = ["completion", "acompletion", "embedding", "aembedding"],
+        List[Literal["completion", "acompletion", "embedding", "aembedding", "atranscription", "transcription"]]
+    ] = ["completion", "acompletion", "embedding", "aembedding", "atranscription", "transcription"],
     **kwargs,
 )
 ```
@@ -215,13 +252,13 @@ Update the Cache params
 
 ```python
 litellm.update_cache(
-    type: Optional[Literal["local", "redis"]] = "local",
+    type: Optional[Literal["local", "redis", "s3", "disk"]] = "local",
     host: Optional[str] = None,
     port: Optional[str] = None,
     password: Optional[str] = None,
     supported_call_types: Optional[
-        List[Literal["completion", "acompletion", "embedding", "aembedding"]]
-    ] = ["completion", "acompletion", "embedding", "aembedding"],
+        List[Literal["completion", "acompletion", "embedding", "aembedding", "atranscription", "transcription"]]
+    ] = ["completion", "acompletion", "embedding", "aembedding", "atranscription", "transcription"],
     **kwargs,
 )
 ```
@@ -276,22 +313,29 @@ cache.get_cache = get_cache
 ```python
 def __init__(
     self,
-    type: Optional[Literal["local", "redis", "s3"]] = "local",
+    type: Optional[Literal["local", "redis", "redis-semantic", "s3", "disk"]] = "local",
     supported_call_types: Optional[
-        List[Literal["completion", "acompletion", "embedding", "aembedding"]]
-    ] = ["completion", "acompletion", "embedding", "aembedding"], # A list of litellm call types to cache for. Defaults to caching for all litellm call types.
-    
+        List[Literal["completion", "acompletion", "embedding", "aembedding", "atranscription", "transcription"]]
+    ] = ["completion", "acompletion", "embedding", "aembedding", "atranscription", "transcription"],
+    ttl: Optional[float] = None,
+    default_in_memory_ttl: Optional[float] = None,
+
     # redis cache params
     host: Optional[str] = None,
     port: Optional[str] = None,
     password: Optional[str] = None,
-
+    namespace: Optional[str] = None,
+    default_in_redis_ttl: Optional[float] = None,
+    similarity_threshold: Optional[float] = None,
+    redis_semantic_cache_use_async=False,
+    redis_semantic_cache_embedding_model="text-embedding-ada-002",
+    redis_flush_size=None,
 
     # s3 Bucket, boto3 configuration
     s3_bucket_name: Optional[str] = None,
     s3_region_name: Optional[str] = None,
     s3_api_version: Optional[str] = None,
-    s3_path: Optional[str] = None, # if you wish to save to a spefic path
+    s3_path: Optional[str] = None, # if you wish to save to a specific path
     s3_use_ssl: Optional[bool] = True,
     s3_verify: Optional[Union[bool, str]] = None,
     s3_endpoint_url: Optional[str] = None,
@@ -299,7 +343,11 @@ def __init__(
     s3_aws_secret_access_key: Optional[str] = None,
     s3_aws_session_token: Optional[str] = None,
     s3_config: Optional[Any] = None,
-    **kwargs,
+
+    # disk cache params
+    disk_cache_dir=None,
+
+    **kwargs
 ):
 ```
 

docs/my-website/docs/caching/local_caching.md

@@ -40,7 +40,7 @@ cache = Cache()
 
 cache.add_cache(cache_key="test-key", result="1234")
 
-cache.get_cache(cache_key="test-key)
+cache.get_cache(cache_key="test-key")
 ```
 
 ## Caching with Streaming 

docs/my-website/docs/completion/input.md

@@ -37,11 +37,12 @@ print(response) # ["max_tokens", "tools", "tool_choice", "stream"]
 
 This is a list of openai params we translate across providers.
 
-This list is constantly being updated.
+Use `litellm.get_supported_openai_params()` for an updated list of params for each model + provider 
 
 | Provider | temperature | max_tokens | top_p | stream | stop | n | presence_penalty | frequency_penalty | functions | function_call | logit_bias | user | response_format | seed | tools | tool_choice | logprobs | top_logprobs | extra_headers | 
 |---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|--|
-|Anthropic| ✅ | ✅ | ✅ | ✅ | ✅ |  |  |   |  |   |
+|Anthropic| ✅ | ✅ | ✅ | ✅ | ✅ |  |  |   |  |  |  |  |  |  | ✅ | ✅ | 
+|Anthropic| ✅ | ✅ | ✅ | ✅ | ✅ |  |  |   |  |  |  |  | ✅ | ✅ | ✅ | ✅ | 
 |OpenAI| ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |✅ | ✅ | ✅ | ✅ |✅ | ✅ | ✅ | ✅ | ✅ |
 |Azure OpenAI| ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |✅ | ✅ | ✅ | ✅ |✅ | ✅ |  |  | ✅ |
 |Replicate | ✅ | ✅ | ✅ | ✅ | ✅ | |  |   |  |   |

docs/my-website/docs/exception_mapping.md

@@ -106,11 +106,12 @@ To see how it's implemented - [check out the code](https://github.com/BerriAI/li
 
 ## Custom mapping list
 
-Base case - we return the original exception.
+Base case - we return `litellm.APIConnectionError` exception (inherits from openai's APIConnectionError exception).
 
 | custom_llm_provider        | Timeout | ContextWindowExceededError | BadRequestError | NotFoundError | ContentPolicyViolationError | AuthenticationError | APIError | RateLimitError | ServiceUnavailableError | PermissionDeniedError | UnprocessableEntityError |
 |----------------------------|---------|----------------------------|------------------|---------------|-----------------------------|---------------------|----------|----------------|-------------------------|-----------------------|-------------------------|
 | openai                     | ✓       | ✓                          | ✓                |               | ✓                           | ✓                   |          |                |                         |                       |                           |
+| watsonx                     |       | | | | | | |✓| | | |
 | text-completion-openai     | ✓       | ✓                          | ✓                |               | ✓                           | ✓                   |          |                |                         |                       |                           |
 | custom_openai              | ✓       | ✓                          | ✓                |               | ✓                           | ✓                   |          |                |                         |                       |                           |
 | openai_compatible_providers| ✓       | ✓                          | ✓                |               | ✓                           | ✓                   |          |                |                         |                       |                           |

docs/my-website/docs/observability/langfuse_integration.md

@@ -213,8 +213,20 @@ chat(messages)
 
 ## Redacting Messages, Response Content from Langfuse Logging 
 
+### Redact Messages and Responses from all Langfuse Logging
+
 Set `litellm.turn_off_message_logging=True` This will prevent the messages and responses from being logged to langfuse, but request metadata will still be logged.
 
+### Redact Messages and Responses from specific Langfuse Logging
+
+In the metadata typically passed for text completion or embedding calls you can set specific keys to mask the messages and responses for this call.
+
+Setting `mask_input` to `True` will mask the input from being logged for this call 
+
+Setting `mask_output` to `True` will make the output from being logged for this call.
+
+Be aware that if you are continuing an existing trace, and you set `update_trace_keys` to include either `input` or `output` and you set the corresponding `mask_input` or `mask_output`, then that trace will have its existing input and/or output replaced with a redacted message.
+
 ## Troubleshooting & Errors
 ### Data not getting logged to Langfuse ? 
 - Ensure you're on the latest version of langfuse `pip install langfuse -U`. The latest version allows litellm to log JSON input/outputs to langfuse

docs/my-website/docs/observability/logfire_integration.md

@@ -0,0 +1,60 @@
+import Image from '@theme/IdealImage';
+
+# Logfire - Logging LLM Input/Output
+
+Logfire is open Source Observability & Analytics for LLM Apps
+Detailed production traces and a granular view on quality, cost and latency
+
+<Image img={require('../../img/logfire.png')} />
+
+:::info
+We want to learn how we can make the callbacks better! Meet the LiteLLM [founders](https://calendly.com/d/4mp-gd3-k5k/berriai-1-1-onboarding-litellm-hosted-version) or
+join our [discord](https://discord.gg/wuPM9dRgDw)
+:::
+
+## Pre-Requisites
+
+Ensure you have run `pip install logfire` for this integration
+
+```shell
+pip install logfire litellm
+```
+
+## Quick Start
+
+Get your Logfire token from [Logfire](https://logfire.pydantic.dev/)
+
+```python
+litellm.success_callback = ["logfire"]
+litellm.failure_callback = ["logfire"] # logs errors to logfire
+```
+
+```python
+# pip install logfire
+import litellm
+import os
+
+# from https://logfire.pydantic.dev/
+os.environ["LOGFIRE_TOKEN"] = ""
+
+# LLM API Keys
+os.environ['OPENAI_API_KEY']=""
+
+# set logfire as a callback, litellm will send the data to logfire
+litellm.success_callback = ["logfire"]
+
+# openai call
+response = litellm.completion(
+  model="gpt-3.5-turbo",
+  messages=[
+    {"role": "user", "content": "Hi 👋 - i'm openai"}
+  ]
+)
+```
+
+## Support & Talk to Founders
+
+- [Schedule Demo 👋](https://calendly.com/d/4mp-gd3-k5k/berriai-1-1-onboarding-litellm-hosted-version)
+- [Community Discord 💭](https://discord.gg/wuPM9dRgDw)
+- Our numbers 📞 +1 (770) 8783-106 / ‭+1 (412) 618-6238‬
+- Our emails ✉️ ishaan@berri.ai / krrish@berri.ai

docs/my-website/docs/providers/openai.md

@@ -20,7 +20,7 @@ os.environ["OPENAI_API_KEY"] = "your-api-key"
 
 # openai call
 response = completion(
-    model = "gpt-3.5-turbo", 
+    model = "gpt-4o", 
     messages=[{ "content": "Hello, how are you?","role": "user"}]
 )
 ```
@@ -163,6 +163,8 @@ os.environ["OPENAI_API_BASE"] = "openaiai-api-base"     # OPTIONAL
 
 | Model Name            | Function Call                                                   |
 |-----------------------|-----------------------------------------------------------------|
+| gpt-4o   | `response = completion(model="gpt-4o", messages=messages)` |
+| gpt-4o-2024-05-13   | `response = completion(model="gpt-4o-2024-05-13", messages=messages)` |
 | gpt-4-turbo   | `response = completion(model="gpt-4-turbo", messages=messages)` |
 | gpt-4-turbo-preview   | `response = completion(model="gpt-4-0125-preview", messages=messages)` |
 | gpt-4-0125-preview    | `response = completion(model="gpt-4-0125-preview", messages=messages)` |

docs/my-website/docs/proxy/alerting.md

@@ -1,13 +1,18 @@
 # 🚨 Alerting 
 
 Get alerts for:
+
 - Hanging LLM api calls
-- Failed LLM api calls
 - Slow LLM api calls
-- Budget Tracking per key/user:
-    - When a User/Key crosses their Budget 
-    - When a User/Key is 15% away from crossing their Budget
+- Failed LLM api calls
+- Budget Tracking per key/user
+- Spend Reports - Weekly & Monthly spend per Team, Tag
 - Failed db read/writes
+- Daily Reports:
+    - **LLM** Top 5 slowest deployments
+    - **LLM** Top 5 deployments with most failed requests
+    - **Spend** Weekly & Monthly spend per Team, Tag
+
 
 ## Quick Start
 
@@ -20,7 +25,8 @@ Get a slack webhook url from https://api.slack.com/messaging/webhooks
 
 ### Step 2: Update config.yaml 
 
-Let's save a bad key to our proxy
+- Set `SLACK_WEBHOOK_URL` in your proxy env to enable Slack alerts.
+- Just for testing purposes, let's save a bad key to our proxy.
 
 ```yaml
 model_list: 
@@ -33,16 +39,23 @@ general_settings:
     alerting: ["slack"]
     alerting_threshold: 300 # sends alerts if requests hang for 5min+ and responses take 5min+ 
 
+environment_variables:
+    SLACK_WEBHOOK_URL: "https://hooks.slack.com/services/<>/<>/<>"
+    SLACK_DAILY_REPORT_FREQUENCY: "86400"  # 24 hours; Optional: defaults to 12 hours
 ```
 
-Set `SLACK_WEBHOOK_URL` in your proxy env
-
-```shell
-SLACK_WEBHOOK_URL: "https://hooks.slack.com/services/<>/<>/<>"
-```
 
 ### Step 3: Start proxy
 
 ```bash
 $ litellm --config /path/to/config.yaml
+```
+
+## Testing Alerting is Setup Correctly
+
+Make a GET request to `/health/services`, expect to see a test slack alert in your provided webhook slack channel
+
+```shell
+curl -X GET 'http://localhost:4000/health/services?service=slack' \
+  -H 'Authorization: Bearer sk-1234'
 ```

docs/my-website/docs/proxy/cost_tracking.md

@@ -1,8 +1,161 @@
-# Cost Tracking - Azure
+import Tabs from '@theme/Tabs';
+import TabItem from '@theme/TabItem';
+
+# 💸 Spend Tracking
+
+Track spend for keys, users, and teams across 100+ LLMs.
+
+## Getting Spend Reports - To Charge Other Teams, API Keys
+
+Use the `/global/spend/report` endpoint to get daily spend per team, with a breakdown of spend per API Key, Model
+
+### Example Request
+
+```shell
+curl -X GET 'http://localhost:4000/global/spend/report?start_date=2023-04-01&end_date=2024-06-30' \
+  -H 'Authorization: Bearer sk-1234'
+```
+
+### Example Response
+<Tabs>
+
+<TabItem value="response" label="Expected Response">
+
+```shell
+[
+    {
+        "group_by_day": "2024-04-30T00:00:00+00:00",
+        "teams": [
+            {
+                "team_name": "Prod Team",
+                "total_spend": 0.0015265,
+                "metadata": [ # see the spend by unique(key + model)
+                    {
+                        "model": "gpt-4",
+                        "spend": 0.00123,
+                        "total_tokens": 28,
+                        "api_key": "88dc28.." # the hashed api key
+                    },
+                    {
+                        "model": "gpt-4",
+                        "spend": 0.00123,
+                        "total_tokens": 28,
+                        "api_key": "a73dc2.." # the hashed api key
+                    },
+                    {
+                        "model": "chatgpt-v-2",
+                        "spend": 0.000214,
+                        "total_tokens": 122,
+                        "api_key": "898c28.." # the hashed api key
+                    },
+                    {
+                        "model": "gpt-3.5-turbo",
+                        "spend": 0.0000825,
+                        "total_tokens": 85,
+                        "api_key": "84dc28.." # the hashed api key
+                    }
+                ]
+            }
+        ]
+    }
+]
+```
+
+
+</TabItem>
+
+<TabItem value="py-script" label="Script to Parse Response (Python)">
+
+```python
+import requests
+url = 'http://localhost:4000/global/spend/report'
+params = {
+    'start_date': '2023-04-01',
+    'end_date': '2024-06-30'
+}
+
+headers = {
+    'Authorization': 'Bearer sk-1234'
+}
+
+# Make the GET request
+response = requests.get(url, headers=headers, params=params)
+spend_report = response.json()
+
+for row in spend_report:
+  date = row["group_by_day"]
+  teams = row["teams"]
+  for team in teams:
+      team_name = team["team_name"]
+      total_spend = team["total_spend"]
+      metadata = team["metadata"]
+
+      print(f"Date: {date}")
+      print(f"Team: {team_name}")
+      print(f"Total Spend: {total_spend}")
+      print("Metadata: ", metadata)
+      print()
+```
+
+Output from script
+```shell
+# Date: 2024-05-11T00:00:00+00:00
+# Team: local_test_team
+# Total Spend: 0.003675099999999999
+# Metadata:  [{'model': 'gpt-3.5-turbo', 'spend': 0.003675099999999999, 'api_key': 'b94d5e0bc3a71a573917fe1335dc0c14728c7016337451af9714924ff3a729db', 'total_tokens': 3105}]
+
+# Date: 2024-05-13T00:00:00+00:00
+# Team: Unassigned Team
+# Total Spend: 3.4e-05
+# Metadata:  [{'model': 'gpt-3.5-turbo', 'spend': 3.4e-05, 'api_key': '9569d13c9777dba68096dea49b0b03e0aaf4d2b65d4030eda9e8a2733c3cd6e0', 'total_tokens': 50}]
+
+# Date: 2024-05-13T00:00:00+00:00
+# Team: central
+# Total Spend: 0.000684
+# Metadata:  [{'model': 'gpt-3.5-turbo', 'spend': 0.000684, 'api_key': '0323facdf3af551594017b9ef162434a9b9a8ca1bbd9ccbd9d6ce173b1015605', 'total_tokens': 498}]
+
+# Date: 2024-05-13T00:00:00+00:00
+# Team: local_test_team
+# Total Spend: 0.0005715000000000001
+# Metadata:  [{'model': 'gpt-3.5-turbo', 'spend': 0.0005715000000000001, 'api_key': 'b94d5e0bc3a71a573917fe1335dc0c14728c7016337451af9714924ff3a729db', 'total_tokens': 423}]
+```
+
+
+</TabItem>
+
+</Tabs>
+
+
+## Reset Team, API Key Spend - MASTER KEY ONLY
+
+Use `/global/spend/reset` if you want to:
+- Reset the Spend for all API Keys, Teams. The `spend` for ALL Teams and Keys in `LiteLLM_TeamTable` and `LiteLLM_VerificationToken` will be set to `spend=0`
+
+- LiteLLM will maintain all the logs in `LiteLLMSpendLogs` for Auditing Purposes
+
+### Request 
+Only the `LITELLM_MASTER_KEY` you set can access this route
+```shell
+curl -X POST \
+  'http://localhost:4000/global/spend/reset' \
+  -H 'Authorization: Bearer sk-1234' \
+  -H 'Content-Type: application/json'
+```
+
+### Expected Responses
+
+```shell
+{"message":"Spend for all API Keys and Teams reset successfully","status":"success"}
+```
+
+
+
+
+## Spend Tracking for Azure
 
 Set base model for cost tracking azure image-gen call
 
-## Image Generation 
+### Image Generation 
 
 ```yaml
 model_list: 
@@ -17,7 +170,7 @@ model_list:
         mode: image_generation
 ```
 
-## Chat Completions / Embeddings
+### Chat Completions / Embeddings
 
 **Problem**: Azure returns `gpt-4` in the response when `azure/gpt-4-1106-preview` is used. This leads to inaccurate cost tracking
 

docs/my-website/docs/proxy/user_keys.md

@@ -365,22 +365,113 @@ curl --location 'http://0.0.0.0:4000/moderations' \
 
 ## Advanced
 
-### (BETA) Batch Completions - pass `model` as List
+### (BETA) Batch Completions - pass multiple models
 
 Use this when you want to send 1 request to N Models
 
 #### Expected Request Format
 
+Pass model as a string of comma separated value of models. Example `"model"="llama3,gpt-3.5-turbo"`
+
 This same request will be sent to the following model groups on the [litellm proxy config.yaml](https://docs.litellm.ai/docs/proxy/configs)
 - `model_name="llama3"`
 - `model_name="gpt-3.5-turbo"` 
 
+<Tabs>
+
+<TabItem value="openai-py" label="OpenAI Python SDK">
+
+
+```python
+import openai
+
+client = openai.OpenAI(api_key="sk-1234", base_url="http://0.0.0.0:4000")
+
+response = client.chat.completions.create(
+    model="gpt-3.5-turbo,llama3",
+    messages=[
+        {"role": "user", "content": "this is a test request, write a short poem"}
+    ],
+)
+
+print(response)
+```
+
+
+
+#### Expected Response Format
+
+Get a list of responses when `model` is passed as a list
+
+```python
+[
+    ChatCompletion(
+        id='chatcmpl-9NoYhS2G0fswot0b6QpoQgmRQMaIf',
+        choices=[
+            Choice(
+                finish_reason='stop',
+                index=0,
+                logprobs=None,
+                message=ChatCompletionMessage(
+                    content='In the depths of my soul, a spark ignites\nA light that shines so pure and bright\nIt dances and leaps, refusing to die\nA flame of hope that reaches the sky\n\nIt warms my heart and fills me with bliss\nA reminder that in darkness, there is light to kiss\nSo I hold onto this fire, this guiding light\nAnd let it lead me through the darkest night.',
+                    role='assistant',
+                    function_call=None,
+                    tool_calls=None
+                )
+            )
+        ],
+        created=1715462919,
+        model='gpt-3.5-turbo-0125',
+        object='chat.completion',
+        system_fingerprint=None,
+        usage=CompletionUsage(
+            completion_tokens=83,
+            prompt_tokens=17,
+            total_tokens=100
+        )
+    ),
+    ChatCompletion(
+        id='chatcmpl-4ac3e982-da4e-486d-bddb-ed1d5cb9c03c',
+        choices=[
+            Choice(
+                finish_reason='stop',
+                index=0,
+                logprobs=None,
+                message=ChatCompletionMessage(
+                    content="A test request, and I'm delighted!\nHere's a short poem, just for you:\n\nMoonbeams dance upon the sea,\nA path of light, for you to see.\nThe stars up high, a twinkling show,\nA night of wonder, for all to know.\n\nThe world is quiet, save the night,\nA peaceful hush, a gentle light.\nThe world is full, of beauty rare,\nA treasure trove, beyond compare.\n\nI hope you enjoyed this little test,\nA poem born, of whimsy and jest.\nLet me know, if there's anything else!",
+                    role='assistant',
+                    function_call=None,
+                    tool_calls=None
+                )
+            )
+        ],
+        created=1715462919,
+        model='groq/llama3-8b-8192',
+        object='chat.completion',
+        system_fingerprint='fp_a2c8d063cb',
+        usage=CompletionUsage(
+            completion_tokens=120,
+            prompt_tokens=20,
+            total_tokens=140
+        )
+    )
+]
+```
+
+
+</TabItem>
+
+<TabItem value="curl" label="Curl">
+
+
+
+
 ```shell
 curl --location 'http://localhost:4000/chat/completions' \
     --header 'Authorization: Bearer sk-1234' \
     --header 'Content-Type: application/json' \
     --data '{
-    "model": ["llama3", "gpt-3.5-turbo"],
+    "model": "llama3,gpt-3.5-turbo",
     "max_tokens": 10,
     "user": "litellm2",
     "messages": [
@@ -393,6 +484,8 @@ curl --location 'http://localhost:4000/chat/completions' \
 ```
 
 
+
+
 #### Expected Response Format
 
 Get a list of responses when `model` is passed as a list
@@ -447,6 +540,11 @@ Get a list of responses when `model` is passed as a list
 ```
 
 
+</TabItem>
+</Tabs>
+
+
+
 
 
 ### Pass User LLM API Keys, Fallbacks

docs/my-website/docs/routing.md

@@ -653,7 +653,9 @@ from litellm import Router
 model_list = [{...}]
 
 router = Router(model_list=model_list, 
-                allowed_fails=1) # cooldown model if it fails > 1 call in a minute. 
+                allowed_fails=1,      # cooldown model if it fails > 1 call in a minute. 
+				cooldown_time=100    # cooldown the deployment for 100 seconds if it num_fails > allowed_fails
+		)
 
 user_message = "Hello, whats the weather in San Francisco??"
 messages = [{"content": user_message, "role": "user"}]
@@ -770,6 +772,8 @@ If the error is a context window exceeded error, fall back to a larger model gro
 
 Fallbacks are done in-order - ["gpt-3.5-turbo, "gpt-4", "gpt-4-32k"], will do 'gpt-3.5-turbo' first, then 'gpt-4', etc.
 
+You can also set 'default_fallbacks', in case a specific model group is misconfigured / bad.
+
 ```python
 from litellm import Router
 
@@ -830,6 +834,7 @@ model_list = [
 
 router = Router(model_list=model_list, 
                 fallbacks=[{"azure/gpt-3.5-turbo": ["gpt-3.5-turbo"]}], 
+				default_fallbacks=["gpt-3.5-turbo-16k"],
                 context_window_fallbacks=[{"azure/gpt-3.5-turbo-context-fallback": ["gpt-3.5-turbo-16k"]}, {"gpt-3.5-turbo": ["gpt-3.5-turbo-16k"]}],
                 set_verbose=True)
 
@@ -1309,10 +1314,11 @@ def __init__(
 	num_retries: int = 0,
 	timeout: Optional[float] = None,
 	default_litellm_params={},  # default params for Router.chat.completion.create
-	fallbacks: List = [],
+	fallbacks: Optional[List] = None,
+	default_fallbacks: Optional[List] = None
 	allowed_fails: Optional[int] = None, # Number of times a deployment can failbefore being added to cooldown
 	cooldown_time: float = 1,  # (seconds) time to cooldown a deployment after failure
-	context_window_fallbacks: List = [],
+	context_window_fallbacks: Optional[List] = None,
 	model_group_alias: Optional[dict] = {},
 	retry_after: int = 0,  # (min) time to wait before retrying a failed request
 	routing_strategy: Literal[

docs/my-website/sidebars.js

@@ -39,6 +39,7 @@ const sidebars = {
         "proxy/demo",
         "proxy/configs",
         "proxy/reliability",
+        "proxy/cost_tracking",
         "proxy/users",
         "proxy/user_keys",
         "proxy/enterprise",
@@ -52,7 +53,6 @@ const sidebars = {
         "proxy/team_based_routing",
         "proxy/customer_routing",
         "proxy/ui",
-        "proxy/cost_tracking",
         "proxy/token_auth",
         {
           type: "category",
@@ -189,7 +189,7 @@ const sidebars = {
         `observability/telemetry`,
       ],
     },
-    "caching/redis_cache",
+    "caching/all_caches",
     {
       type: "category",
       label: "Tutorials",

enterprise/utils.py

@@ -1,6 +1,7 @@
 # Enterprise Proxy Util Endpoints
 from litellm._logging import verbose_logger
 import collections
+from datetime import datetime
 
 
 async def get_spend_by_tags(start_date=None, end_date=None, prisma_client=None):
@@ -18,26 +19,33 @@ async def get_spend_by_tags(start_date=None, end_date=None, prisma_client=None):
     return response
 
 
-async def ui_get_spend_by_tags(start_date=None, end_date=None, prisma_client=None):
-    response = await prisma_client.db.query_raw(
-        """
+async def ui_get_spend_by_tags(start_date: str, end_date: str, prisma_client):
+
+    sql_query = """
         SELECT
         jsonb_array_elements_text(request_tags) AS individual_request_tag,
         DATE(s."startTime") AS spend_date,
         COUNT(*) AS log_count,
         SUM(spend) AS total_spend
         FROM "LiteLLM_SpendLogs" s
-        WHERE s."startTime" >= current_date - interval '30 days'
+        WHERE
+            DATE(s."startTime") >= $1::date
+            AND DATE(s."startTime") <= $2::date
         GROUP BY individual_request_tag, spend_date
-        ORDER BY spend_date;
-        """
+        ORDER BY spend_date
+        LIMIT 100;
+    """
+    response = await prisma_client.db.query_raw(
+        sql_query,
+        start_date,
+        end_date,
     )
 
     # print("tags - spend")
     # print(response)
     # Bar Chart 1 - Spend per tag - Top 10 tags by spend
-    total_spend_per_tag = collections.defaultdict(float)
-    total_requests_per_tag = collections.defaultdict(int)
+    total_spend_per_tag: collections.defaultdict = collections.defaultdict(float)
+    total_requests_per_tag: collections.defaultdict = collections.defaultdict(int)
     for row in response:
         tag_name = row["individual_request_tag"]
         tag_spend = row["total_spend"]
@@ -49,15 +57,18 @@ async def ui_get_spend_by_tags(start_date=None, end_date=None, prisma_client=Non
     # convert to ui format
     ui_tags = []
     for tag in sorted_tags:
+        current_spend = tag[1]
+        if current_spend is not None and isinstance(current_spend, float):
+            current_spend = round(current_spend, 4)
         ui_tags.append(
             {
                 "name": tag[0],
-                "value": tag[1],
+                "spend": current_spend,
                 "log_count": total_requests_per_tag[tag[0]],
             }
         )
 
-    return {"top_10_tags": ui_tags}
+    return {"spend_per_tag": ui_tags}
 
 
 async def view_spend_logs_from_clickhouse(

litellm/__init__.py

@@ -219,6 +219,7 @@ max_end_user_budget: Optional[float] = None
 #### RELIABILITY ####
 request_timeout: Optional[float] = 6000
 num_retries: Optional[int] = None  # per model endpoint
+default_fallbacks: Optional[List] = None
 fallbacks: Optional[List] = None
 context_window_fallbacks: Optional[List] = None
 allowed_fails: int = 0
@@ -406,69 +407,69 @@ replicate_models: List = [
 ]
 
 clarifai_models: List = [
-    'clarifai/meta.Llama-3.Llama-3-8B-Instruct', 
-    'clarifai/gcp.generate.gemma-1_1-7b-it', 
-    'clarifai/mistralai.completion.mixtral-8x22B', 
-    'clarifai/cohere.generate.command-r-plus', 
-    'clarifai/databricks.drbx.dbrx-instruct', 
-    'clarifai/mistralai.completion.mistral-large',
-    'clarifai/mistralai.completion.mistral-medium', 
-    'clarifai/mistralai.completion.mistral-small', 
-    'clarifai/mistralai.completion.mixtral-8x7B-Instruct-v0_1',
-    'clarifai/gcp.generate.gemma-2b-it',
-    'clarifai/gcp.generate.gemma-7b-it', 
-    'clarifai/deci.decilm.deciLM-7B-instruct',
-    'clarifai/mistralai.completion.mistral-7B-Instruct',
-    'clarifai/gcp.generate.gemini-pro', 
-    'clarifai/anthropic.completion.claude-v1', 
-    'clarifai/anthropic.completion.claude-instant-1_2', 
-    'clarifai/anthropic.completion.claude-instant', 
-    'clarifai/anthropic.completion.claude-v2',
-    'clarifai/anthropic.completion.claude-2_1',
-    'clarifai/meta.Llama-2.codeLlama-70b-Python',
-    'clarifai/meta.Llama-2.codeLlama-70b-Instruct',
-    'clarifai/openai.completion.gpt-3_5-turbo-instruct', 
-    'clarifai/meta.Llama-2.llama2-7b-chat',
-    'clarifai/meta.Llama-2.llama2-13b-chat', 
-    'clarifai/meta.Llama-2.llama2-70b-chat', 
-    'clarifai/openai.chat-completion.gpt-4-turbo',
-    'clarifai/microsoft.text-generation.phi-2',
-    'clarifai/meta.Llama-2.llama2-7b-chat-vllm',
-    'clarifai/upstage.solar.solar-10_7b-instruct', 
-    'clarifai/openchat.openchat.openchat-3_5-1210',
-    'clarifai/togethercomputer.stripedHyena.stripedHyena-Nous-7B',
-    'clarifai/gcp.generate.text-bison', 
-    'clarifai/meta.Llama-2.llamaGuard-7b', 
-    'clarifai/fblgit.una-cybertron.una-cybertron-7b-v2',
-    'clarifai/openai.chat-completion.GPT-4', 
-    'clarifai/openai.chat-completion.GPT-3_5-turbo', 
-    'clarifai/ai21.complete.Jurassic2-Grande',
-    'clarifai/ai21.complete.Jurassic2-Grande-Instruct',
-    'clarifai/ai21.complete.Jurassic2-Jumbo-Instruct',
-    'clarifai/ai21.complete.Jurassic2-Jumbo', 
-    'clarifai/ai21.complete.Jurassic2-Large', 
-    'clarifai/cohere.generate.cohere-generate-command',
-    'clarifai/wizardlm.generate.wizardCoder-Python-34B',
-    'clarifai/wizardlm.generate.wizardLM-70B', 
-    'clarifai/tiiuae.falcon.falcon-40b-instruct', 
-    'clarifai/togethercomputer.RedPajama.RedPajama-INCITE-7B-Chat',
-    'clarifai/gcp.generate.code-gecko', 
-    'clarifai/gcp.generate.code-bison', 
-    'clarifai/mistralai.completion.mistral-7B-OpenOrca',
-    'clarifai/mistralai.completion.openHermes-2-mistral-7B', 
-    'clarifai/wizardlm.generate.wizardLM-13B',
-    'clarifai/huggingface-research.zephyr.zephyr-7B-alpha', 
-    'clarifai/wizardlm.generate.wizardCoder-15B', 
-    'clarifai/microsoft.text-generation.phi-1_5', 
-    'clarifai/databricks.Dolly-v2.dolly-v2-12b', 
-    'clarifai/bigcode.code.StarCoder', 
-    'clarifai/salesforce.xgen.xgen-7b-8k-instruct', 
-    'clarifai/mosaicml.mpt.mpt-7b-instruct',  
-    'clarifai/anthropic.completion.claude-3-opus',
-    'clarifai/anthropic.completion.claude-3-sonnet',
-    'clarifai/gcp.generate.gemini-1_5-pro',
-    'clarifai/gcp.generate.imagen-2',
-    'clarifai/salesforce.blip.general-english-image-caption-blip-2', 
+    "clarifai/meta.Llama-3.Llama-3-8B-Instruct",
+    "clarifai/gcp.generate.gemma-1_1-7b-it",
+    "clarifai/mistralai.completion.mixtral-8x22B",
+    "clarifai/cohere.generate.command-r-plus",
+    "clarifai/databricks.drbx.dbrx-instruct",
+    "clarifai/mistralai.completion.mistral-large",
+    "clarifai/mistralai.completion.mistral-medium",
+    "clarifai/mistralai.completion.mistral-small",
+    "clarifai/mistralai.completion.mixtral-8x7B-Instruct-v0_1",
+    "clarifai/gcp.generate.gemma-2b-it",
+    "clarifai/gcp.generate.gemma-7b-it",
+    "clarifai/deci.decilm.deciLM-7B-instruct",
+    "clarifai/mistralai.completion.mistral-7B-Instruct",
+    "clarifai/gcp.generate.gemini-pro",
+    "clarifai/anthropic.completion.claude-v1",
+    "clarifai/anthropic.completion.claude-instant-1_2",
+    "clarifai/anthropic.completion.claude-instant",
+    "clarifai/anthropic.completion.claude-v2",
+    "clarifai/anthropic.completion.claude-2_1",
+    "clarifai/meta.Llama-2.codeLlama-70b-Python",
+    "clarifai/meta.Llama-2.codeLlama-70b-Instruct",
+    "clarifai/openai.completion.gpt-3_5-turbo-instruct",
+    "clarifai/meta.Llama-2.llama2-7b-chat",
+    "clarifai/meta.Llama-2.llama2-13b-chat",
+    "clarifai/meta.Llama-2.llama2-70b-chat",
+    "clarifai/openai.chat-completion.gpt-4-turbo",
+    "clarifai/microsoft.text-generation.phi-2",
+    "clarifai/meta.Llama-2.llama2-7b-chat-vllm",
+    "clarifai/upstage.solar.solar-10_7b-instruct",
+    "clarifai/openchat.openchat.openchat-3_5-1210",
+    "clarifai/togethercomputer.stripedHyena.stripedHyena-Nous-7B",
+    "clarifai/gcp.generate.text-bison",
+    "clarifai/meta.Llama-2.llamaGuard-7b",
+    "clarifai/fblgit.una-cybertron.una-cybertron-7b-v2",
+    "clarifai/openai.chat-completion.GPT-4",
+    "clarifai/openai.chat-completion.GPT-3_5-turbo",
+    "clarifai/ai21.complete.Jurassic2-Grande",
+    "clarifai/ai21.complete.Jurassic2-Grande-Instruct",
+    "clarifai/ai21.complete.Jurassic2-Jumbo-Instruct",
+    "clarifai/ai21.complete.Jurassic2-Jumbo",
+    "clarifai/ai21.complete.Jurassic2-Large",
+    "clarifai/cohere.generate.cohere-generate-command",
+    "clarifai/wizardlm.generate.wizardCoder-Python-34B",
+    "clarifai/wizardlm.generate.wizardLM-70B",
+    "clarifai/tiiuae.falcon.falcon-40b-instruct",
+    "clarifai/togethercomputer.RedPajama.RedPajama-INCITE-7B-Chat",
+    "clarifai/gcp.generate.code-gecko",
+    "clarifai/gcp.generate.code-bison",
+    "clarifai/mistralai.completion.mistral-7B-OpenOrca",
+    "clarifai/mistralai.completion.openHermes-2-mistral-7B",
+    "clarifai/wizardlm.generate.wizardLM-13B",
+    "clarifai/huggingface-research.zephyr.zephyr-7B-alpha",
+    "clarifai/wizardlm.generate.wizardCoder-15B",
+    "clarifai/microsoft.text-generation.phi-1_5",
+    "clarifai/databricks.Dolly-v2.dolly-v2-12b",
+    "clarifai/bigcode.code.StarCoder",
+    "clarifai/salesforce.xgen.xgen-7b-8k-instruct",
+    "clarifai/mosaicml.mpt.mpt-7b-instruct",
+    "clarifai/anthropic.completion.claude-3-opus",
+    "clarifai/anthropic.completion.claude-3-sonnet",
+    "clarifai/gcp.generate.gemini-1_5-pro",
+    "clarifai/gcp.generate.imagen-2",
+    "clarifai/salesforce.blip.general-english-image-caption-blip-2",
 ]
 
 
@@ -755,7 +756,7 @@ from .llms.bedrock import (
     AmazonMistralConfig,
     AmazonBedrockGlobalConfig,
 )
-from .llms.openai import OpenAIConfig, OpenAITextCompletionConfig
+from .llms.openai import OpenAIConfig, OpenAITextCompletionConfig, MistralConfig
 from .llms.azure import AzureOpenAIConfig, AzureOpenAIError
 from .llms.watsonx import IBMWatsonXAIConfig
 from .main import *  # type: ignore

litellm/caching.py

@@ -373,11 +373,12 @@ class RedisCache(BaseCache):
                         print_verbose(
                             f"Set ASYNC Redis Cache PIPELINE: key: {cache_key}\nValue {cache_value}\nttl={ttl}"
                         )
+                        json_cache_value = json.dumps(cache_value)
                         # Set the value with a TTL if it's provided.
                         if ttl is not None:
-                            pipe.setex(cache_key, ttl, json.dumps(cache_value))
+                            pipe.setex(cache_key, ttl, json_cache_value)
                         else:
-                            pipe.set(cache_key, json.dumps(cache_value))
+                            pipe.set(cache_key, json_cache_value)
                     # Execute the pipeline and return the results.
                     results = await pipe.execute()
 
@@ -810,9 +811,7 @@ class RedisSemanticCache(BaseCache):
 
         # get the prompt
         messages = kwargs["messages"]
-        prompt = ""
-        for message in messages:
-            prompt += message["content"]
+        prompt = "".join(message["content"] for message in messages)
 
         # create an embedding for prompt
         embedding_response = litellm.embedding(
@@ -847,9 +846,7 @@ class RedisSemanticCache(BaseCache):
 
         # get the messages
         messages = kwargs["messages"]
-        prompt = ""
-        for message in messages:
-            prompt += message["content"]
+        prompt = "".join(message["content"] for message in messages)
 
         # convert to embedding
         embedding_response = litellm.embedding(
@@ -909,9 +906,7 @@ class RedisSemanticCache(BaseCache):
 
         # get the prompt
         messages = kwargs["messages"]
-        prompt = ""
-        for message in messages:
-            prompt += message["content"]
+        prompt = "".join(message["content"] for message in messages)
         # create an embedding for prompt
         router_model_names = (
             [m["model_name"] for m in llm_model_list]
@@ -964,9 +959,7 @@ class RedisSemanticCache(BaseCache):
 
         # get the messages
         messages = kwargs["messages"]
-        prompt = ""
-        for message in messages:
-            prompt += message["content"]
+        prompt = "".join(message["content"] for message in messages)
 
         router_model_names = (
             [m["model_name"] for m in llm_model_list]
@@ -1448,7 +1441,7 @@ class DualCache(BaseCache):
 class Cache:
     def __init__(
         self,
-        type: Optional[Literal["local", "redis", "redis-semantic", "s3"]] = "local",
+        type: Optional[Literal["local", "redis", "redis-semantic", "s3", "disk"]] = "local",
         host: Optional[str] = None,
         port: Optional[str] = None,
         password: Optional[str] = None,
@@ -1491,13 +1484,14 @@ class Cache:
         redis_semantic_cache_use_async=False,
         redis_semantic_cache_embedding_model="text-embedding-ada-002",
         redis_flush_size=None,
+        disk_cache_dir=None,
         **kwargs,
     ):
         """
         Initializes the cache based on the given type.
 
         Args:
-            type (str, optional): The type of cache to initialize. Can be "local", "redis", "redis-semantic", or "s3". Defaults to "local".
+            type (str, optional): The type of cache to initialize. Can be "local", "redis", "redis-semantic", "s3" or "disk". Defaults to "local".
             host (str, optional): The host address for the Redis cache. Required if type is "redis".
             port (int, optional): The port number for the Redis cache. Required if type is "redis".
             password (str, optional): The password for the Redis cache. Required if type is "redis".
@@ -1543,6 +1537,8 @@ class Cache:
                 s3_path=s3_path,
                 **kwargs,
             )
+        elif type == "disk":
+            self.cache = DiskCache(disk_cache_dir=disk_cache_dir)
         if "cache" not in litellm.input_callback:
             litellm.input_callback.append("cache")
         if "cache" not in litellm.success_callback:
@@ -1914,8 +1910,86 @@ class Cache:
             await self.cache.disconnect()
 
 
+class DiskCache(BaseCache):
+    def __init__(self, disk_cache_dir: Optional[str] = None):
+        import diskcache as dc
+
+        # if users don't provider one, use the default litellm cache
+        if disk_cache_dir is None:
+            self.disk_cache = dc.Cache(".litellm_cache")
+        else:
+            self.disk_cache = dc.Cache(disk_cache_dir)
+
+    def set_cache(self, key, value, **kwargs):
+        print_verbose("DiskCache: set_cache")
+        if "ttl" in kwargs:
+            self.disk_cache.set(key, value, expire=kwargs["ttl"])
+        else:
+            self.disk_cache.set(key, value)
+
+    async def async_set_cache(self, key, value, **kwargs):
+        self.set_cache(key=key, value=value, **kwargs)
+
+    async def async_set_cache_pipeline(self, cache_list, ttl=None):
+        for cache_key, cache_value in cache_list:
+            if ttl is not None:
+                self.set_cache(key=cache_key, value=cache_value, ttl=ttl)
+            else:
+                self.set_cache(key=cache_key, value=cache_value)
+
+    def get_cache(self, key, **kwargs):
+        original_cached_response = self.disk_cache.get(key)
+        if original_cached_response:
+            try:
+                cached_response = json.loads(original_cached_response)
+            except:
+                cached_response = original_cached_response
+            return cached_response
+        return None
+
+    def batch_get_cache(self, keys: list, **kwargs):
+        return_val = []
+        for k in keys:
+            val = self.get_cache(key=k, **kwargs)
+            return_val.append(val)
+        return return_val
+
+    def increment_cache(self, key, value: int, **kwargs) -> int:
+        # get the value
+        init_value = self.get_cache(key=key) or 0
+        value = init_value + value
+        self.set_cache(key, value, **kwargs)
+        return value
+
+    async def async_get_cache(self, key, **kwargs):
+        return self.get_cache(key=key, **kwargs)
+
+    async def async_batch_get_cache(self, keys: list, **kwargs):
+        return_val = []
+        for k in keys:
+            val = self.get_cache(key=k, **kwargs)
+            return_val.append(val)
+        return return_val
+
+    async def async_increment(self, key, value: int, **kwargs) -> int:
+        # get the value
+        init_value = await self.async_get_cache(key=key) or 0
+        value = init_value + value
+        await self.async_set_cache(key, value, **kwargs)
+        return value
+
+    def flush_cache(self):
+        self.disk_cache.clear()
+
+    async def disconnect(self):
+        pass
+
+    def delete_cache(self, key):
+        self.disk_cache.pop(key)
+
+
 def enable_cache(
-    type: Optional[Literal["local", "redis", "s3"]] = "local",
+    type: Optional[Literal["local", "redis", "s3", "disk"]] = "local",
     host: Optional[str] = None,
     port: Optional[str] = None,
     password: Optional[str] = None,
@@ -1944,7 +2018,7 @@ def enable_cache(
     Enable cache with the specified configuration.
 
     Args:
-        type (Optional[Literal["local", "redis"]]): The type of cache to enable. Defaults to "local".
+        type (Optional[Literal["local", "redis", "s3", "disk"]]): The type of cache to enable. Defaults to "local".
         host (Optional[str]): The host address of the cache server. Defaults to None.
         port (Optional[str]): The port number of the cache server. Defaults to None.
         password (Optional[str]): The password for the cache server. Defaults to None.
@@ -1980,7 +2054,7 @@ def enable_cache(
 
 
 def update_cache(
-    type: Optional[Literal["local", "redis"]] = "local",
+    type: Optional[Literal["local", "redis", "s3", "disk"]] = "local",
     host: Optional[str] = None,
     port: Optional[str] = None,
     password: Optional[str] = None,
@@ -2009,7 +2083,7 @@ def update_cache(
     Update the cache for LiteLLM.
 
     Args:
-        type (Optional[Literal["local", "redis"]]): The type of cache. Defaults to "local".
+        type (Optional[Literal["local", "redis", "s3", "disk"]]): The type of cache. Defaults to "local".
         host (Optional[str]): The host of the cache. Defaults to None.
         port (Optional[str]): The port of the cache. Defaults to None.
         password (Optional[str]): The password for the cache. Defaults to None.

litellm/integrations/langfuse.py

@@ -322,6 +322,8 @@ class LangFuseLogger:
             existing_trace_id = clean_metadata.pop("existing_trace_id", None)
             update_trace_keys = clean_metadata.pop("update_trace_keys", [])
             debug = clean_metadata.pop("debug_langfuse", None)
+            mask_input = clean_metadata.pop("mask_input", False)
+            mask_output = clean_metadata.pop("mask_output", False)
 
             if trace_name is None and existing_trace_id is None:
                 # just log `litellm-{call_type}` as the trace name
@@ -349,15 +351,15 @@ class LangFuseLogger:
 
                 # Special keys that are found in the function arguments and not the metadata
                 if "input" in update_trace_keys:
-                    trace_params["input"] = input
+                    trace_params["input"] = input if not mask_input else "redacted-by-litellm"
                 if "output" in update_trace_keys:
-                    trace_params["output"] = output
+                    trace_params["output"] = output if not mask_output else "redacted-by-litellm"
             else:  # don't overwrite an existing trace
                 trace_params = {
                     "id": trace_id,
                     "name": trace_name,
                     "session_id": session_id,
-                    "input": input,
+                    "input": input if not mask_input else "redacted-by-litellm",
                     "version": clean_metadata.pop(
                         "trace_version", clean_metadata.get("version", None)
                     ),  # If provided just version, it will applied to the trace as well, if applied a trace version it will take precedence
@@ -373,7 +375,7 @@ class LangFuseLogger:
                 if level == "ERROR":
                     trace_params["status_message"] = output
                 else:
-                    trace_params["output"] = output
+                    trace_params["output"] = output if not mask_output else "redacted-by-litellm"
 
             if debug == True or (isinstance(debug, str) and debug.lower() == "true"):
                 if "metadata" in trace_params:
@@ -463,8 +465,8 @@ class LangFuseLogger:
                 "end_time": end_time,
                 "model": kwargs["model"],
                 "model_parameters": optional_params,
-                "input": input,
-                "output": output,
+                "input": input if not mask_input else "redacted-by-litellm",
+                "output": output if not mask_output else "redacted-by-litellm",
                 "usage": usage,
                 "metadata": clean_metadata,
                 "level": level,

litellm/integrations/logfire_logger.py

@@ -0,0 +1,178 @@
+#### What this does ####
+#    On success + failure, log events to Logfire
+
+import dotenv, os
+
+dotenv.load_dotenv()  # Loading env variables using dotenv
+import traceback
+import uuid
+from litellm._logging import print_verbose, verbose_logger
+
+from enum import Enum
+from typing import Any, Dict, NamedTuple
+from typing_extensions import LiteralString
+
+
+class SpanConfig(NamedTuple):
+    message_template: LiteralString
+    span_data: Dict[str, Any]
+
+
+class LogfireLevel(str, Enum):
+    INFO = "info"
+    ERROR = "error"
+
+
+class LogfireLogger:
+    # Class variables or attributes
+    def __init__(self):
+        try:
+            verbose_logger.debug(f"in init logfire logger")
+            import logfire
+
+            # only setting up logfire if we are sending to logfire
+            # in testing, we don't want to send to logfire
+            if logfire.DEFAULT_LOGFIRE_INSTANCE.config.send_to_logfire:
+                logfire.configure(token=os.getenv("LOGFIRE_TOKEN"))
+        except Exception as e:
+            print_verbose(f"Got exception on init logfire client {str(e)}")
+            raise e
+
+    def _get_span_config(self, payload) -> SpanConfig:
+        if (
+            payload["call_type"] == "completion"
+            or payload["call_type"] == "acompletion"
+        ):
+            return SpanConfig(
+                message_template="Chat Completion with {request_data[model]!r}",
+                span_data={"request_data": payload},
+            )
+        elif (
+            payload["call_type"] == "embedding" or payload["call_type"] == "aembedding"
+        ):
+            return SpanConfig(
+                message_template="Embedding Creation with {request_data[model]!r}",
+                span_data={"request_data": payload},
+            )
+        elif (
+            payload["call_type"] == "image_generation"
+            or payload["call_type"] == "aimage_generation"
+        ):
+            return SpanConfig(
+                message_template="Image Generation with {request_data[model]!r}",
+                span_data={"request_data": payload},
+            )
+        else:
+            return SpanConfig(
+                message_template="Litellm Call with {request_data[model]!r}",
+                span_data={"request_data": payload},
+            )
+
+    async def _async_log_event(
+        self,
+        kwargs,
+        response_obj,
+        start_time,
+        end_time,
+        print_verbose,
+        level: LogfireLevel,
+    ):
+        self.log_event(
+            kwargs=kwargs,
+            response_obj=response_obj,
+            start_time=start_time,
+            end_time=end_time,
+            print_verbose=print_verbose,
+            level=level,
+        )
+
+    def log_event(
+        self,
+        kwargs,
+        start_time,
+        end_time,
+        print_verbose,
+        level: LogfireLevel,
+        response_obj,
+    ):
+        try:
+            import logfire
+
+            verbose_logger.debug(
+                f"logfire Logging - Enters logging function for model {kwargs}"
+            )
+
+            if not response_obj:
+                response_obj = {}
+            litellm_params = kwargs.get("litellm_params", {})
+            metadata = (
+                litellm_params.get("metadata", {}) or {}
+            )  # if litellm_params['metadata'] == None
+            messages = kwargs.get("messages")
+            optional_params = kwargs.get("optional_params", {})
+            call_type = kwargs.get("call_type", "completion")
+            cache_hit = kwargs.get("cache_hit", False)
+            usage = response_obj.get("usage", {})
+            id = response_obj.get("id", str(uuid.uuid4()))
+            try:
+                response_time = (end_time - start_time).total_seconds()
+            except:
+                response_time = None
+
+            # Clean Metadata before logging - never log raw metadata
+            # the raw metadata can contain circular references which leads to infinite recursion
+            # we clean out all extra litellm metadata params before logging
+            clean_metadata = {}
+            if isinstance(metadata, dict):
+                for key, value in metadata.items():
+                    # clean litellm metadata before logging
+                    if key in [
+                        "endpoint",
+                        "caching_groups",
+                        "previous_models",
+                    ]:
+                        continue
+                    else:
+                        clean_metadata[key] = value
+
+            # Build the initial payload
+            payload = {
+                "id": id,
+                "call_type": call_type,
+                "cache_hit": cache_hit,
+                "startTime": start_time,
+                "endTime": end_time,
+                "responseTime (seconds)": response_time,
+                "model": kwargs.get("model", ""),
+                "user": kwargs.get("user", ""),
+                "modelParameters": optional_params,
+                "spend": kwargs.get("response_cost", 0),
+                "messages": messages,
+                "response": response_obj,
+                "usage": usage,
+                "metadata": clean_metadata,
+            }
+            logfire_openai = logfire.with_settings(custom_scope_suffix="openai")
+            message_template, span_data = self._get_span_config(payload)
+            if level == LogfireLevel.INFO:
+                logfire_openai.info(
+                    message_template,
+                    **span_data,
+                )
+            elif level == LogfireLevel.ERROR:
+                logfire_openai.error(
+                    message_template,
+                    **span_data,
+                    _exc_info=True,
+                )
+            print_verbose(f"\ndd Logger - Logging payload = {payload}")
+
+            print_verbose(
+                f"Logfire Layer Logging - final response object: {response_obj}"
+            )
+        except Exception as e:
+            traceback.print_exc()
+            verbose_logger.debug(
+                f"Logfire Layer Error - {str(e)}\n{traceback.format_exc()}"
+            )
+            pass

litellm/integrations/slack_alerting.py

@@ -31,7 +31,8 @@ class LiteLLMBase(BaseModel):
 
 
 class SlackAlertingArgs(LiteLLMBase):
-    daily_report_frequency: int = 12 * 60 * 60  # 12 hours
+    default_daily_report_frequency: int = 12 * 60 * 60  # 12 hours
+    daily_report_frequency: int = int(os.getenv("SLACK_DAILY_REPORT_FREQUENCY", default_daily_report_frequency))
     report_check_interval: int = 5 * 60  # 5 minutes
 
 
@@ -76,16 +77,14 @@ class SlackAlerting(CustomLogger):
         internal_usage_cache: Optional[DualCache] = None,
         alerting_threshold: float = 300,  # threshold for slow / hanging llm responses (in seconds)
         alerting: Optional[List] = [],
-        alert_types: Optional[
-            List[
-                Literal[
-                    "llm_exceptions",
-                    "llm_too_slow",
-                    "llm_requests_hanging",
-                    "budget_alerts",
-                    "db_exceptions",
-                    "daily_reports",
-                ]
+        alert_types: List[
+            Literal[
+                "llm_exceptions",
+                "llm_too_slow",
+                "llm_requests_hanging",
+                "budget_alerts",
+                "db_exceptions",
+                "daily_reports",
             ]
         ] = [
             "llm_exceptions",
@@ -240,6 +239,8 @@ class SlackAlerting(CustomLogger):
                 end_time=end_time,
             )
         )
+        if litellm.turn_off_message_logging:
+            messages = "Message not logged. `litellm.turn_off_message_logging=True`."
         request_info = f"\nRequest Model: `{model}`\nAPI Base: `{api_base}`\nMessages: `{messages}`"
         slow_message = f"`Responses are slow - {round(time_difference_float,2)}s response time > Alerting threshold: {self.alerting_threshold}s`"
         if time_difference_float > self.alerting_threshold:
@@ -346,8 +347,9 @@ class SlackAlerting(CustomLogger):
 
         all_none = True
         for val in combined_metrics_values:
-            if val is not None:
+            if val is not None and val > 0:
                 all_none = False
+                break
 
         if all_none:
             return False
@@ -365,13 +367,14 @@ class SlackAlerting(CustomLogger):
             for value in failed_request_values
         ]
 
-        ## Get the indices of top 5 keys with the highest numerical values (ignoring None values)
+        ## Get the indices of top 5 keys with the highest numerical values (ignoring None and 0 values)
         top_5_failed = sorted(
             range(len(replaced_failed_values)),
             key=lambda i: replaced_failed_values[i],
             reverse=True,
         )[:5]
-
+        top_5_failed = [index for index in top_5_failed if replaced_failed_values[index] > 0]
+        
         # find top 5 slowest
         # Replace None values with a placeholder value (-1 in this case)
         placeholder_value = 0
@@ -380,17 +383,20 @@ class SlackAlerting(CustomLogger):
             for value in latency_values
         ]
 
-        # Get the indices of top 5 values with the highest numerical values (ignoring None values)
+        # Get the indices of top 5 values with the highest numerical values (ignoring None and 0 values)
         top_5_slowest = sorted(
             range(len(replaced_slowest_values)),
             key=lambda i: replaced_slowest_values[i],
             reverse=True,
         )[:5]
+        top_5_slowest = [index for index in top_5_slowest if replaced_slowest_values[index] > 0]
 
         # format alert -> return the litellm model name + api base
         message = f"\n\nHere are today's key metrics 📈: \n\n"
 
-        message += "\n\n*❗️ Top 5 Deployments with Most Failed Requests:*\n\n"
+        message += "\n\n*❗️ Top Deployments with Most Failed Requests:*\n\n"
+        if not top_5_failed:
+            message += "\tNone\n"
         for i in range(len(top_5_failed)):
             key = failed_request_keys[top_5_failed[i]].split(":")[0]
             _deployment = router.get_model_info(key)
@@ -410,7 +416,9 @@ class SlackAlerting(CustomLogger):
             value = replaced_failed_values[top_5_failed[i]]
             message += f"\t{i+1}. Deployment: `{deployment_name}`, Failed Requests: `{value}`,  API Base: `{api_base}`\n"
 
-        message += "\n\n*😅 Top 5 Slowest Deployments:*\n\n"
+        message += "\n\n*😅 Top Slowest Deployments:*\n\n"
+        if not top_5_slowest:
+            message += "\tNone\n"
         for i in range(len(top_5_slowest)):
             key = latency_keys[top_5_slowest[i]].split(":")[0]
             _deployment = router.get_model_info(key)
@@ -462,6 +470,11 @@ class SlackAlerting(CustomLogger):
                 messages = messages[:100]
             except:
                 messages = ""
+
+            if litellm.turn_off_message_logging:
+                messages = (
+                    "Message not logged. `litellm.turn_off_message_logging=True`."
+                )
             request_info = f"\nRequest Model: `{model}`\nMessages: `{messages}`"
         else:
             request_info = ""
@@ -812,14 +825,6 @@ Model Info:
                     updated_at=litellm.utils.get_utc_datetime(),
                 )
             )
-        if "llm_exceptions" in self.alert_types:
-            original_exception = kwargs.get("exception", None)
-
-            await self.send_alert(
-                message="LLM API Failure - " + str(original_exception),
-                level="High",
-                alert_type="llm_exceptions",
-            )
 
     async def _run_scheduler_helper(self, llm_router) -> bool:
         """
@@ -883,3 +888,99 @@ Model Info:
                 )  # shuffle to prevent collisions
                 await asyncio.sleep(interval)
         return
+
+    async def send_weekly_spend_report(self):
+        """ """
+        try:
+            from litellm.proxy.proxy_server import _get_spend_report_for_time_range
+
+            todays_date = datetime.datetime.now().date()
+            week_before = todays_date - datetime.timedelta(days=7)
+
+            weekly_spend_per_team, weekly_spend_per_tag = (
+                await _get_spend_report_for_time_range(
+                    start_date=week_before.strftime("%Y-%m-%d"),
+                    end_date=todays_date.strftime("%Y-%m-%d"),
+                )
+            )
+
+            _weekly_spend_message = f"*💸 Weekly Spend Report for `{week_before.strftime('%m-%d-%Y')} - {todays_date.strftime('%m-%d-%Y')}` *\n"
+
+            if weekly_spend_per_team is not None:
+                _weekly_spend_message += "\n*Team Spend Report:*\n"
+                for spend in weekly_spend_per_team:
+                    _team_spend = spend["total_spend"]
+                    _team_spend = float(_team_spend)
+                    # round to 4 decimal places
+                    _team_spend = round(_team_spend, 4)
+                    _weekly_spend_message += (
+                        f"Team: `{spend['team_alias']}` | Spend: `${_team_spend}`\n"
+                    )
+
+            if weekly_spend_per_tag is not None:
+                _weekly_spend_message += "\n*Tag Spend Report:*\n"
+                for spend in weekly_spend_per_tag:
+                    _tag_spend = spend["total_spend"]
+                    _tag_spend = float(_tag_spend)
+                    # round to 4 decimal places
+                    _tag_spend = round(_tag_spend, 4)
+                    _weekly_spend_message += f"Tag: `{spend['individual_request_tag']}` | Spend: `${_tag_spend}`\n"
+
+            await self.send_alert(
+                message=_weekly_spend_message,
+                level="Low",
+                alert_type="daily_reports",
+            )
+        except Exception as e:
+            verbose_proxy_logger.error("Error sending weekly spend report", e)
+
+    async def send_monthly_spend_report(self):
+        """ """
+        try:
+            from calendar import monthrange
+
+            from litellm.proxy.proxy_server import _get_spend_report_for_time_range
+
+            todays_date = datetime.datetime.now().date()
+            first_day_of_month = todays_date.replace(day=1)
+            _, last_day_of_month = monthrange(todays_date.year, todays_date.month)
+            last_day_of_month = first_day_of_month + datetime.timedelta(
+                days=last_day_of_month - 1
+            )
+
+            monthly_spend_per_team, monthly_spend_per_tag = (
+                await _get_spend_report_for_time_range(
+                    start_date=first_day_of_month.strftime("%Y-%m-%d"),
+                    end_date=last_day_of_month.strftime("%Y-%m-%d"),
+                )
+            )
+
+            _spend_message = f"*💸 Monthly Spend Report for `{first_day_of_month.strftime('%m-%d-%Y')} - {last_day_of_month.strftime('%m-%d-%Y')}` *\n"
+
+            if monthly_spend_per_team is not None:
+                _spend_message += "\n*Team Spend Report:*\n"
+                for spend in monthly_spend_per_team:
+                    _team_spend = spend["total_spend"]
+                    _team_spend = float(_team_spend)
+                    # round to 4 decimal places
+                    _team_spend = round(_team_spend, 4)
+                    _spend_message += (
+                        f"Team: `{spend['team_alias']}` | Spend: `${_team_spend}`\n"
+                    )
+
+            if monthly_spend_per_tag is not None:
+                _spend_message += "\n*Tag Spend Report:*\n"
+                for spend in monthly_spend_per_tag:
+                    _tag_spend = spend["total_spend"]
+                    _tag_spend = float(_tag_spend)
+                    # round to 4 decimal places
+                    _tag_spend = round(_tag_spend, 4)
+                    _spend_message += f"Tag: `{spend['individual_request_tag']}` | Spend: `${_tag_spend}`\n"
+
+            await self.send_alert(
+                message=_spend_message,
+                level="Low",
+                alert_type="daily_reports",
+            )
+        except Exception as e:
+            verbose_proxy_logger.error("Error sending weekly spend report", e)

litellm/llms/ollama_chat.py

@@ -300,7 +300,7 @@ def get_ollama_response(
         model_response["choices"][0]["message"] = message
         model_response["choices"][0]["finish_reason"] = "tool_calls"
     else:
-        model_response["choices"][0]["message"] = response_json["message"]
+        model_response["choices"][0]["message"]["content"] = response_json["message"]["content"]
     model_response["created"] = int(time.time())
     model_response["model"] = "ollama/" + model
     prompt_tokens = response_json.get("prompt_eval_count", litellm.token_counter(messages=messages))  # type: ignore
@@ -484,7 +484,7 @@ async def ollama_acompletion(
                 model_response["choices"][0]["message"] = message
                 model_response["choices"][0]["finish_reason"] = "tool_calls"
             else:
-                model_response["choices"][0]["message"] = response_json["message"]
+                model_response["choices"][0]["message"]["content"] = response_json["message"]["content"]
 
             model_response["created"] = int(time.time())
             model_response["model"] = "ollama_chat/" + data["model"]

litellm/llms/openai.py

@@ -53,6 +53,113 @@ class OpenAIError(Exception):
         )  # Call the base class constructor with the parameters it needs
 
 
+class MistralConfig:
+    """
+    Reference: https://docs.mistral.ai/api/
+
+    The class `MistralConfig` provides configuration for the Mistral's Chat API interface. Below are the parameters:
+
+    - `temperature` (number or null): Defines the sampling temperature to use, varying between 0 and 2. API Default - 0.7.
+
+    - `top_p` (number or null): An alternative to sampling with temperature, used for nucleus sampling. API Default - 1.
+
+    - `max_tokens` (integer or null): This optional parameter helps to set the maximum number of tokens to generate in the chat completion. API Default - null.
+
+    - `tools` (list or null): A list of available tools for the model. Use this to specify functions for which the model can generate JSON inputs.
+
+    - `tool_choice` (string - 'auto'/'any'/'none' or null): Specifies if/how functions are called. If set to none the model won't call a function and will generate a message instead. If set to auto the model can choose to either generate a message or call a function. If set to any the model is forced to call a function. Default - 'auto'.
+
+    - `random_seed` (integer or null): The seed to use for random sampling. If set, different calls will generate deterministic results.
+
+    - `safe_prompt` (boolean): Whether to inject a safety prompt before all conversations. API Default - 'false'.
+
+    - `response_format` (object or null): An object specifying the format that the model must output. Setting to { "type": "json_object" } enables JSON mode, which guarantees the message the model generates is in JSON. When using JSON mode you MUST also instruct the model to produce JSON yourself with a system or a user message.
+    """
+
+    temperature: Optional[int] = None
+    top_p: Optional[int] = None
+    max_tokens: Optional[int] = None
+    tools: Optional[list] = None
+    tool_choice: Optional[Literal["auto", "any", "none"]] = None
+    random_seed: Optional[int] = None
+    safe_prompt: Optional[bool] = None
+    response_format: Optional[dict] = None
+
+    def __init__(
+        self,
+        temperature: Optional[int] = None,
+        top_p: Optional[int] = None,
+        max_tokens: Optional[int] = None,
+        tools: Optional[list] = None,
+        tool_choice: Optional[Literal["auto", "any", "none"]] = None,
+        random_seed: Optional[int] = None,
+        safe_prompt: Optional[bool] = None,
+        response_format: Optional[dict] = None,
+    ) -> None:
+        locals_ = locals()
+        for key, value in locals_.items():
+            if key != "self" and value is not None:
+                setattr(self.__class__, key, value)
+
+    @classmethod
+    def get_config(cls):
+        return {
+            k: v
+            for k, v in cls.__dict__.items()
+            if not k.startswith("__")
+            and not isinstance(
+                v,
+                (
+                    types.FunctionType,
+                    types.BuiltinFunctionType,
+                    classmethod,
+                    staticmethod,
+                ),
+            )
+            and v is not None
+        }
+
+    def get_supported_openai_params(self):
+        return [
+            "stream",
+            "temperature",
+            "top_p",
+            "max_tokens",
+            "tools",
+            "tool_choice",
+            "seed",
+            "response_format",
+        ]
+
+    def _map_tool_choice(self, tool_choice: str) -> str:
+        if tool_choice == "auto" or tool_choice == "none":
+            return tool_choice
+        elif tool_choice == "required":
+            return "any"
+        else:  # openai 'tool_choice' object param not supported by Mistral API
+            return "any"
+
+    def map_openai_params(self, non_default_params: dict, optional_params: dict):
+        for param, value in non_default_params.items():
+            if param == "max_tokens":
+                optional_params["max_tokens"] = value
+            if param == "tools":
+                optional_params["tools"] = value
+            if param == "stream" and value == True:
+                optional_params["stream"] = value
+            if param == "temperature":
+                optional_params["temperature"] = value
+            if param == "top_p":
+                optional_params["top_p"] = value
+            if param == "tool_choice" and isinstance(value, str):
+                optional_params["tool_choice"] = self._map_tool_choice(
+                    tool_choice=value
+                )
+            if param == "seed":
+                optional_params["extra_body"] = {"random_seed": value}
+        return optional_params
+
+
 class OpenAIConfig:
     """
     Reference: https://platform.openai.com/docs/api-reference/chat/create
@@ -1327,8 +1434,8 @@ class OpenAIAssistantsAPI(BaseLLM):
             client=client,
         )
 
-        thread_message: OpenAIMessage = openai_client.beta.threads.messages.create(
-            thread_id, **message_data
+        thread_message: OpenAIMessage = openai_client.beta.threads.messages.create(  # type: ignore
+            thread_id, **message_data  # type: ignore
         )
 
         response_obj: Optional[OpenAIMessage] = None
@@ -1458,7 +1565,7 @@ class OpenAIAssistantsAPI(BaseLLM):
             client=client,
         )
 
-        response = openai_client.beta.threads.runs.create_and_poll(
+        response = openai_client.beta.threads.runs.create_and_poll(  # type: ignore
             thread_id=thread_id,
             assistant_id=assistant_id,
             additional_instructions=additional_instructions,

litellm/llms/vertex_ai.py

@@ -867,6 +867,8 @@ async def async_completion(
     Add support for acompletion calls for gemini-pro
     """
     try:
+        import proto  # type: ignore
+
         if mode == "vision":
             print_verbose("\nMaking VertexAI Gemini Pro/Vision Call")
             print_verbose(f"\nProcessing input messages = {messages}")
@@ -901,9 +903,21 @@ async def async_completion(
             ):
                 function_call = response.candidates[0].content.parts[0].function_call
                 args_dict = {}
-                for k, v in function_call.args.items():
-                    args_dict[k] = v
-                args_str = json.dumps(args_dict)
+
+                # Check if it's a RepeatedComposite instance
+                for key, val in function_call.args.items():
+                    if isinstance(
+                        val, proto.marshal.collections.repeated.RepeatedComposite
+                    ):
+                        # If so, convert to list
+                        args_dict[key] = [v for v in val]
+                    else:
+                        args_dict[key] = val
+
+                try:
+                    args_str = json.dumps(args_dict)
+                except Exception as e:
+                    raise VertexAIError(status_code=422, message=str(e))
                 message = litellm.Message(
                     content=None,
                     tool_calls=[

litellm/main.py

@@ -9,12 +9,12 @@
 
 import os, openai, sys, json, inspect, uuid, datetime, threading
 from typing import Any, Literal, Union, BinaryIO
+from typing_extensions import overload
 from functools import partial
 import dotenv, traceback, random, asyncio, time, contextvars
 from copy import deepcopy
 import httpx
 import litellm
-
 from ._logging import verbose_logger
 from litellm import (  # type: ignore
     client,
@@ -727,7 +727,6 @@ def completion(
 
         ### REGISTER CUSTOM MODEL PRICING -- IF GIVEN ###
         if input_cost_per_token is not None and output_cost_per_token is not None:
-            print_verbose(f"Registering model={model} in model cost map")
             litellm.register_model(
                 {
                     f"{custom_llm_provider}/{model}": {
@@ -849,6 +848,10 @@ def completion(
             proxy_server_request=proxy_server_request,
             preset_cache_key=preset_cache_key,
             no_log=no_log,
+            input_cost_per_second=input_cost_per_second,
+            input_cost_per_token=input_cost_per_token,
+            output_cost_per_second=output_cost_per_second,
+            output_cost_per_token=output_cost_per_token,
         )
         logging.update_environment_variables(
             model=model,

litellm/model_prices_and_context_window_backup.json

@@ -9,6 +9,30 @@
         "mode": "chat",
         "supports_function_calling": true
     },
+    "gpt-4o": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.000005,
+        "output_cost_per_token": 0.000015,
+        "litellm_provider": "openai",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true
+    },
+    "gpt-4o-2024-05-13": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.000005,
+        "output_cost_per_token": 0.000015,
+        "litellm_provider": "openai",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true
+    },
     "gpt-4-turbo-preview": {
         "max_tokens": 4096,
         "max_input_tokens": 128000,
@@ -3366,4 +3390,4 @@
         "mode": "embedding"
     }
 
-}
+}

litellm/proxy/_experimental/out/index.html

@@ -1 +1 @@
-<!DOCTYPE html><html id="__next_error__"><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width, initial-scale=1"/><link rel="preload" as="script" fetchPriority="low" href="/ui/_next/static/chunks/webpack-de9c0fadf6a94b3b.js" crossorigin=""/><script src="/ui/_next/static/chunks/fd9d1056-f960ab1e6d32b002.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/69-04708d7d4a17c1ee.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/main-app-9b4fb13a7db53edf.js" async="" crossorigin=""></script><title>LiteLLM Dashboard</title><meta name="description" content="LiteLLM Proxy Admin UI"/><link rel="icon" href="/ui/favicon.ico" type="image/x-icon" sizes="16x16"/><meta name="next-size-adjust"/><script src="/ui/_next/static/chunks/polyfills-c67a75d1b6f99dc8.js" crossorigin="" noModule=""></script></head><body><script src="/ui/_next/static/chunks/webpack-de9c0fadf6a94b3b.js" crossorigin="" async=""></script><script>(self.__next_f=self.__next_f||[]).push([0]);self.__next_f.push([2,null])</script><script>self.__next_f.push([1,"1:HL[\"/ui/_next/static/media/c9a5bc6a7c948fb0-s.p.woff2\",\"font\",{\"crossOrigin\":\"\",\"type\":\"font/woff2\"}]\n2:HL[\"/ui/_next/static/css/f04e46b02318b660.css\",\"style\",{\"crossOrigin\":\"\"}]\n0:\"$L3\"\n"])</script><script>self.__next_f.push([1,"4:I[47690,[],\"\"]\n6:I[77831,[],\"\"]\n7:I[7926,[\"936\",\"static/chunks/2f6dbc85-052c4579f80d66ae.js\",\"884\",\"static/chunks/884-7576ee407a2ecbe6.js\",\"931\",\"static/chunks/app/page-e6190351ac8da62a.js\"],\"\"]\n8:I[5613,[],\"\"]\n9:I[31778,[],\"\"]\nb:I[48955,[],\"\"]\nc:[]\n"])</script><script>self.__next_f.push([1,"3:[[[\"$\",\"link\",\"0\",{\"rel\":\"stylesheet\",\"href\":\"/ui/_next/static/css/f04e46b02318b660.css\",\"precedence\":\"next\",\"crossOrigin\":\"\"}]],[\"$\",\"$L4\",null,{\"buildId\":\"84BZ5uERcn4DsO4_POsLl\",\"assetPrefix\":\"/ui\",\"initialCanonicalUrl\":\"/\",\"initialTree\":[\"\",{\"children\":[\"__PAGE__\",{}]},\"$undefined\",\"$undefined\",true],\"initialSeedData\":[\"\",{\"children\":[\"__PAGE__\",{},[\"$L5\",[\"$\",\"$L6\",null,{\"propsForComponent\":{\"params\":{}},\"Component\":\"$7\",\"isStaticGeneration\":true}],null]]},[null,[\"$\",\"html\",null,{\"lang\":\"en\",\"children\":[\"$\",\"body\",null,{\"className\":\"__className_c23dc8\",\"children\":[\"$\",\"$L8\",null,{\"parallelRouterKey\":\"children\",\"segmentPath\":[\"children\"],\"loading\":\"$undefined\",\"loadingStyles\":\"$undefined\",\"loadingScripts\":\"$undefined\",\"hasLoading\":false,\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L9\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":[[\"$\",\"title\",null,{\"children\":\"404: This page could not be found.\"}],[\"$\",\"div\",null,{\"style\":{\"fontFamily\":\"system-ui,\\\"Segoe UI\\\",Roboto,Helvetica,Arial,sans-serif,\\\"Apple Color Emoji\\\",\\\"Segoe UI Emoji\\\"\",\"height\":\"100vh\",\"textAlign\":\"center\",\"display\":\"flex\",\"flexDirection\":\"column\",\"alignItems\":\"center\",\"justifyContent\":\"center\"},\"children\":[\"$\",\"div\",null,{\"children\":[[\"$\",\"style\",null,{\"dangerouslySetInnerHTML\":{\"__html\":\"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}\"}}],[\"$\",\"h1\",null,{\"className\":\"next-error-h1\",\"style\":{\"display\":\"inline-block\",\"margin\":\"0 20px 0 0\",\"padding\":\"0 23px 0 0\",\"fontSize\":24,\"fontWeight\":500,\"verticalAlign\":\"top\",\"lineHeight\":\"49px\"},\"children\":\"404\"}],[\"$\",\"div\",null,{\"style\":{\"display\":\"inline-block\"},\"children\":[\"$\",\"h2\",null,{\"style\":{\"fontSize\":14,\"fontWeight\":400,\"lineHeight\":\"49px\",\"margin\":0},\"children\":\"This page could not be found.\"}]}]]}]}]],\"notFoundStyles\":[],\"styles\":null}]}]}],null]],\"initialHead\":[false,\"$La\"],\"globalErrorComponent\":\"$b\",\"missingSlots\":\"$Wc\"}]]\n"])</script><script>self.__next_f.push([1,"a:[[\"$\",\"meta\",\"0\",{\"name\":\"viewport\",\"content\":\"width=device-width, initial-scale=1\"}],[\"$\",\"meta\",\"1\",{\"charSet\":\"utf-8\"}],[\"$\",\"title\",\"2\",{\"children\":\"LiteLLM Dashboard\"}],[\"$\",\"meta\",\"3\",{\"name\":\"description\",\"content\":\"LiteLLM Proxy Admin UI\"}],[\"$\",\"link\",\"4\",{\"rel\":\"icon\",\"href\":\"/ui/favicon.ico\",\"type\":\"image/x-icon\",\"sizes\":\"16x16\"}],[\"$\",\"meta\",\"5\",{\"name\":\"next-size-adjust\"}]]\n5:null\n"])</script><script>self.__next_f.push([1,""])</script></body></html>
+<!DOCTYPE html><html id="__next_error__"><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width, initial-scale=1"/><link rel="preload" as="script" fetchPriority="low" href="/ui/_next/static/chunks/webpack-de9c0fadf6a94b3b.js" crossorigin=""/><script src="/ui/_next/static/chunks/fd9d1056-f960ab1e6d32b002.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/69-04708d7d4a17c1ee.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/main-app-9b4fb13a7db53edf.js" async="" crossorigin=""></script><title>LiteLLM Dashboard</title><meta name="description" content="LiteLLM Proxy Admin UI"/><link rel="icon" href="/ui/favicon.ico" type="image/x-icon" sizes="16x16"/><meta name="next-size-adjust"/><script src="/ui/_next/static/chunks/polyfills-c67a75d1b6f99dc8.js" crossorigin="" noModule=""></script></head><body><script src="/ui/_next/static/chunks/webpack-de9c0fadf6a94b3b.js" crossorigin="" async=""></script><script>(self.__next_f=self.__next_f||[]).push([0]);self.__next_f.push([2,null])</script><script>self.__next_f.push([1,"1:HL[\"/ui/_next/static/media/c9a5bc6a7c948fb0-s.p.woff2\",\"font\",{\"crossOrigin\":\"\",\"type\":\"font/woff2\"}]\n2:HL[\"/ui/_next/static/css/f04e46b02318b660.css\",\"style\",{\"crossOrigin\":\"\"}]\n0:\"$L3\"\n"])</script><script>self.__next_f.push([1,"4:I[47690,[],\"\"]\n6:I[77831,[],\"\"]\n7:I[7926,[\"936\",\"static/chunks/2f6dbc85-052c4579f80d66ae.js\",\"884\",\"static/chunks/884-7576ee407a2ecbe6.js\",\"931\",\"static/chunks/app/page-6a39771cacf75ea6.js\"],\"\"]\n8:I[5613,[],\"\"]\n9:I[31778,[],\"\"]\nb:I[48955,[],\"\"]\nc:[]\n"])</script><script>self.__next_f.push([1,"3:[[[\"$\",\"link\",\"0\",{\"rel\":\"stylesheet\",\"href\":\"/ui/_next/static/css/f04e46b02318b660.css\",\"precedence\":\"next\",\"crossOrigin\":\"\"}]],[\"$\",\"$L4\",null,{\"buildId\":\"obp5wqVSVDMiDTC414cR8\",\"assetPrefix\":\"/ui\",\"initialCanonicalUrl\":\"/\",\"initialTree\":[\"\",{\"children\":[\"__PAGE__\",{}]},\"$undefined\",\"$undefined\",true],\"initialSeedData\":[\"\",{\"children\":[\"__PAGE__\",{},[\"$L5\",[\"$\",\"$L6\",null,{\"propsForComponent\":{\"params\":{}},\"Component\":\"$7\",\"isStaticGeneration\":true}],null]]},[null,[\"$\",\"html\",null,{\"lang\":\"en\",\"children\":[\"$\",\"body\",null,{\"className\":\"__className_c23dc8\",\"children\":[\"$\",\"$L8\",null,{\"parallelRouterKey\":\"children\",\"segmentPath\":[\"children\"],\"loading\":\"$undefined\",\"loadingStyles\":\"$undefined\",\"loadingScripts\":\"$undefined\",\"hasLoading\":false,\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L9\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":[[\"$\",\"title\",null,{\"children\":\"404: This page could not be found.\"}],[\"$\",\"div\",null,{\"style\":{\"fontFamily\":\"system-ui,\\\"Segoe UI\\\",Roboto,Helvetica,Arial,sans-serif,\\\"Apple Color Emoji\\\",\\\"Segoe UI Emoji\\\"\",\"height\":\"100vh\",\"textAlign\":\"center\",\"display\":\"flex\",\"flexDirection\":\"column\",\"alignItems\":\"center\",\"justifyContent\":\"center\"},\"children\":[\"$\",\"div\",null,{\"children\":[[\"$\",\"style\",null,{\"dangerouslySetInnerHTML\":{\"__html\":\"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}\"}}],[\"$\",\"h1\",null,{\"className\":\"next-error-h1\",\"style\":{\"display\":\"inline-block\",\"margin\":\"0 20px 0 0\",\"padding\":\"0 23px 0 0\",\"fontSize\":24,\"fontWeight\":500,\"verticalAlign\":\"top\",\"lineHeight\":\"49px\"},\"children\":\"404\"}],[\"$\",\"div\",null,{\"style\":{\"display\":\"inline-block\"},\"children\":[\"$\",\"h2\",null,{\"style\":{\"fontSize\":14,\"fontWeight\":400,\"lineHeight\":\"49px\",\"margin\":0},\"children\":\"This page could not be found.\"}]}]]}]}]],\"notFoundStyles\":[],\"styles\":null}]}]}],null]],\"initialHead\":[false,\"$La\"],\"globalErrorComponent\":\"$b\",\"missingSlots\":\"$Wc\"}]]\n"])</script><script>self.__next_f.push([1,"a:[[\"$\",\"meta\",\"0\",{\"name\":\"viewport\",\"content\":\"width=device-width, initial-scale=1\"}],[\"$\",\"meta\",\"1\",{\"charSet\":\"utf-8\"}],[\"$\",\"title\",\"2\",{\"children\":\"LiteLLM Dashboard\"}],[\"$\",\"meta\",\"3\",{\"name\":\"description\",\"content\":\"LiteLLM Proxy Admin UI\"}],[\"$\",\"link\",\"4\",{\"rel\":\"icon\",\"href\":\"/ui/favicon.ico\",\"type\":\"image/x-icon\",\"sizes\":\"16x16\"}],[\"$\",\"meta\",\"5\",{\"name\":\"next-size-adjust\"}]]\n5:null\n"])</script><script>self.__next_f.push([1,""])</script></body></html>

litellm/proxy/_experimental/out/index.txt

@@ -1,7 +1,7 @@
 2:I[77831,[],""]
-3:I[7926,["936","static/chunks/2f6dbc85-052c4579f80d66ae.js","884","static/chunks/884-7576ee407a2ecbe6.js","931","static/chunks/app/page-e6190351ac8da62a.js"],""]
+3:I[7926,["936","static/chunks/2f6dbc85-052c4579f80d66ae.js","884","static/chunks/884-7576ee407a2ecbe6.js","931","static/chunks/app/page-6a39771cacf75ea6.js"],""]
 4:I[5613,[],""]
 5:I[31778,[],""]
-0:["84BZ5uERcn4DsO4_POsLl",[[["",{"children":["__PAGE__",{}]},"$undefined","$undefined",true],["",{"children":["__PAGE__",{},["$L1",["$","$L2",null,{"propsForComponent":{"params":{}},"Component":"$3","isStaticGeneration":true}],null]]},[null,["$","html",null,{"lang":"en","children":["$","body",null,{"className":"__className_c23dc8","children":["$","$L4",null,{"parallelRouterKey":"children","segmentPath":["children"],"loading":"$undefined","loadingStyles":"$undefined","loadingScripts":"$undefined","hasLoading":false,"error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L5",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":[["$","title",null,{"children":"404: This page could not be found."}],["$","div",null,{"style":{"fontFamily":"system-ui,\"Segoe UI\",Roboto,Helvetica,Arial,sans-serif,\"Apple Color Emoji\",\"Segoe UI Emoji\"","height":"100vh","textAlign":"center","display":"flex","flexDirection":"column","alignItems":"center","justifyContent":"center"},"children":["$","div",null,{"children":[["$","style",null,{"dangerouslySetInnerHTML":{"__html":"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}"}}],["$","h1",null,{"className":"next-error-h1","style":{"display":"inline-block","margin":"0 20px 0 0","padding":"0 23px 0 0","fontSize":24,"fontWeight":500,"verticalAlign":"top","lineHeight":"49px"},"children":"404"}],["$","div",null,{"style":{"display":"inline-block"},"children":["$","h2",null,{"style":{"fontSize":14,"fontWeight":400,"lineHeight":"49px","margin":0},"children":"This page could not be found."}]}]]}]}]],"notFoundStyles":[],"styles":null}]}]}],null]],[[["$","link","0",{"rel":"stylesheet","href":"/ui/_next/static/css/f04e46b02318b660.css","precedence":"next","crossOrigin":""}]],"$L6"]]]]
+0:["obp5wqVSVDMiDTC414cR8",[[["",{"children":["__PAGE__",{}]},"$undefined","$undefined",true],["",{"children":["__PAGE__",{},["$L1",["$","$L2",null,{"propsForComponent":{"params":{}},"Component":"$3","isStaticGeneration":true}],null]]},[null,["$","html",null,{"lang":"en","children":["$","body",null,{"className":"__className_c23dc8","children":["$","$L4",null,{"parallelRouterKey":"children","segmentPath":["children"],"loading":"$undefined","loadingStyles":"$undefined","loadingScripts":"$undefined","hasLoading":false,"error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L5",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":[["$","title",null,{"children":"404: This page could not be found."}],["$","div",null,{"style":{"fontFamily":"system-ui,\"Segoe UI\",Roboto,Helvetica,Arial,sans-serif,\"Apple Color Emoji\",\"Segoe UI Emoji\"","height":"100vh","textAlign":"center","display":"flex","flexDirection":"column","alignItems":"center","justifyContent":"center"},"children":["$","div",null,{"children":[["$","style",null,{"dangerouslySetInnerHTML":{"__html":"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}"}}],["$","h1",null,{"className":"next-error-h1","style":{"display":"inline-block","margin":"0 20px 0 0","padding":"0 23px 0 0","fontSize":24,"fontWeight":500,"verticalAlign":"top","lineHeight":"49px"},"children":"404"}],["$","div",null,{"style":{"display":"inline-block"},"children":["$","h2",null,{"style":{"fontSize":14,"fontWeight":400,"lineHeight":"49px","margin":0},"children":"This page could not be found."}]}]]}]}]],"notFoundStyles":[],"styles":null}]}]}],null]],[[["$","link","0",{"rel":"stylesheet","href":"/ui/_next/static/css/f04e46b02318b660.css","precedence":"next","crossOrigin":""}]],"$L6"]]]]
 6:[["$","meta","0",{"name":"viewport","content":"width=device-width, initial-scale=1"}],["$","meta","1",{"charSet":"utf-8"}],["$","title","2",{"children":"LiteLLM Dashboard"}],["$","meta","3",{"name":"description","content":"LiteLLM Proxy Admin UI"}],["$","link","4",{"rel":"icon","href":"/ui/favicon.ico","type":"image/x-icon","sizes":"16x16"}],["$","meta","5",{"name":"next-size-adjust"}]]
 1:null

litellm/proxy/_super_secret_config.yaml

@@ -8,15 +8,28 @@ model_list:
     base_model: text-embedding-ada-002
     mode: embedding
   model_name: text-embedding-ada-002
+- model_name: gpt-3.5-turbo-012
+  litellm_params:
+    model: gpt-3.5-turbo
+    api_base: http://0.0.0.0:8080
+    api_key: "" 
+- model_name: gpt-3.5-turbo-0125-preview
+  litellm_params:
+    model: azure/chatgpt-v-2
+    api_key: os.environ/AZURE_API_KEY
+    api_base: os.environ/AZURE_API_BASE
+    input_cost_per_token: 0.0
+    output_cost_per_token: 0.0
 
 router_settings:
   redis_host: redis
   # redis_password: <your redis password>
   redis_port: 6379
+  enable_pre_call_checks: true
 
 litellm_settings:
   set_verbose: True
-  enable_preview_features: true
+  fallbacks: [{"gpt-3.5-turbo-012": ["gpt-3.5-turbo-0125-preview"]}]
   # service_callback: ["prometheus_system"]
   # success_callback: ["prometheus"]
   # failure_callback: ["prometheus"]
@@ -25,4 +38,5 @@ general_settings:
   enable_jwt_auth: True
   disable_reset_budget: True
   proxy_batch_write_at: 60 # 👈 Frequency of batch writing logs to server (in seconds)
-  routing_strategy: simple-shuffle # Literal["simple-shuffle", "least-busy", "usage-based-routing","latency-based-routing"], default="simple-shuffle"
+  routing_strategy: simple-shuffle # Literal["simple-shuffle", "least-busy", "usage-based-routing","latency-based-routing"], default="simple-shuffle"
+  alerting: ["slack"]

litellm/proxy/_types.py

@@ -1,11 +1,20 @@
-from pydantic import BaseModel, Extra, Field, root_validator, Json, validator
-from dataclasses import fields
+from pydantic import ConfigDict, BaseModel, Field, root_validator, Json
 import enum
 from typing import Optional, List, Union, Dict, Literal, Any
 from datetime import datetime
-import uuid, json, sys, os
+import uuid
+import json
 from litellm.types.router import UpdateRouterConfig
 
+try:
+    from pydantic import model_validator  # pydantic v2
+except ImportError:
+    from pydantic import root_validator  # pydantic v1
+
+    def model_validator(mode):
+        pre = mode == "before"
+        return root_validator(pre=pre)
+
 
 def hash_token(token: str):
     import hashlib
@@ -35,8 +44,9 @@ class LiteLLMBase(BaseModel):
             # if using pydantic v1
             return self.__fields_set__
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
 
 class LiteLLM_UpperboundKeyGenerateParams(LiteLLMBase):
@@ -79,9 +89,15 @@ class LiteLLMRoutes(enum.Enum):
         "/v1/models",
     ]
 
+    # NOTE: ROUTES ONLY FOR MASTER KEY - only the Master Key should be able to Reset Spend
+    master_key_only_routes: List = [
+        "/global/spend/reset",
+    ]
+
     info_routes: List = [
         "/key/info",
         "/team/info",
+        "/team/list",
         "/user/info",
         "/model/info",
         "/v2/model/info",
@@ -110,6 +126,7 @@ class LiteLLMRoutes(enum.Enum):
         "/team/new",
         "/team/update",
         "/team/delete",
+        "/team/list",
         "/team/info",
         "/team/block",
         "/team/unblock",
@@ -189,7 +206,12 @@ class LiteLLM_JWTAuth(LiteLLMBase):
             "spend_tracking_routes",
             "global_spend_tracking_routes",
         ]
-    ] = ["management_routes", "spend_tracking_routes", "global_spend_tracking_routes"]
+    ] = [
+        "management_routes",
+        "spend_tracking_routes",
+        "global_spend_tracking_routes",
+        "info_routes",
+    ]
     team_jwt_scope: str = "litellm_team"
     team_id_jwt_field: str = "client_id"
     team_allowed_routes: List[
@@ -222,7 +244,7 @@ class LiteLLMPromptInjectionParams(LiteLLMBase):
     llm_api_system_prompt: Optional[str] = None
     llm_api_fail_call_string: Optional[str] = None
 
-    @root_validator(pre=True)
+    @model_validator(mode="before")
     def check_llm_api_params(cls, values):
         llm_api_check = values.get("llm_api_check")
         if llm_api_check is True:
@@ -280,8 +302,9 @@ class ProxyChatCompletionRequest(LiteLLMBase):
     deployment_id: Optional[str] = None
     request_timeout: Optional[int] = None
 
-    class Config:
-        extra = "allow"  # allow params not defined here, these fall in litellm.completion(**kwargs)
+    model_config = ConfigDict(
+        extra = "allow",  # allow params not defined here, these fall in litellm.completion(**kwargs)
+    )
 
 
 class ModelInfoDelete(LiteLLMBase):
@@ -308,11 +331,12 @@ class ModelInfo(LiteLLMBase):
         ]
     ]
 
-    class Config:
-        extra = Extra.allow  # Allow extra fields
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        extra = "allow",  # Allow extra fields
+        protected_namespaces = (),
+    )
 
-    @root_validator(pre=True)
+    @model_validator(mode="before")
     def set_model_info(cls, values):
         if values.get("id") is None:
             values.update({"id": str(uuid.uuid4())})
@@ -338,10 +362,11 @@ class ModelParams(LiteLLMBase):
     litellm_params: dict
     model_info: ModelInfo
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
-    @root_validator(pre=True)
+    @model_validator(mode="before")
     def set_model_info(cls, values):
         if values.get("model_info") is None:
             values.update({"model_info": ModelInfo()})
@@ -377,8 +402,9 @@ class GenerateKeyRequest(GenerateRequestBase):
         {}
     )  # {"gpt-4": 5.0, "gpt-3.5-turbo": 5.0}, defaults to {}
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
 
 class GenerateKeyResponse(GenerateKeyRequest):
@@ -388,7 +414,7 @@ class GenerateKeyResponse(GenerateKeyRequest):
     user_id: Optional[str] = None
     token_id: Optional[str] = None
 
-    @root_validator(pre=True)
+    @model_validator(mode="before")
     def set_model_info(cls, values):
         if values.get("token") is not None:
             values.update({"key": values.get("token")})
@@ -428,8 +454,9 @@ class LiteLLM_ModelTable(LiteLLMBase):
     created_by: str
     updated_by: str
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
 
 class NewUserRequest(GenerateKeyRequest):
@@ -457,7 +484,7 @@ class UpdateUserRequest(GenerateRequestBase):
     user_role: Optional[str] = None
     max_budget: Optional[float] = None
 
-    @root_validator(pre=True)
+    @model_validator(mode="before")
     def check_user_info(cls, values):
         if values.get("user_id") is None and values.get("user_email") is None:
             raise ValueError("Either user id or user email must be provided")
@@ -477,7 +504,7 @@ class NewEndUserRequest(LiteLLMBase):
         None  # if no equivalent model in allowed region - default all requests to this model
     )
 
-    @root_validator(pre=True)
+    @model_validator(mode="before")
     def check_user_info(cls, values):
         if values.get("max_budget") is not None and values.get("budget_id") is not None:
             raise ValueError("Set either 'max_budget' or 'budget_id', not both.")
@@ -490,7 +517,7 @@ class Member(LiteLLMBase):
     user_id: Optional[str] = None
     user_email: Optional[str] = None
 
-    @root_validator(pre=True)
+    @model_validator(mode="before")
     def check_user_info(cls, values):
         if values.get("user_id") is None and values.get("user_email") is None:
             raise ValueError("Either user id or user email must be provided")
@@ -515,8 +542,9 @@ class TeamBase(LiteLLMBase):
 class NewTeamRequest(TeamBase):
     model_aliases: Optional[dict] = None
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
 
 class GlobalEndUsersSpend(LiteLLMBase):
@@ -535,7 +563,7 @@ class TeamMemberDeleteRequest(LiteLLMBase):
     user_id: Optional[str] = None
     user_email: Optional[str] = None
 
-    @root_validator(pre=True)
+    @model_validator(mode="before")
     def check_user_info(cls, values):
         if values.get("user_id") is None and values.get("user_email") is None:
             raise ValueError("Either user id or user email must be provided")
@@ -569,10 +597,11 @@ class LiteLLM_TeamTable(TeamBase):
     budget_reset_at: Optional[datetime] = None
     model_id: Optional[int] = None
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
-    @root_validator(pre=True)
+    @model_validator(mode="before")
     def set_model_info(cls, values):
         dict_fields = [
             "metadata",
@@ -608,8 +637,9 @@ class LiteLLM_BudgetTable(LiteLLMBase):
     model_max_budget: Optional[dict] = None
     budget_duration: Optional[str] = None
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
 
 class NewOrganizationRequest(LiteLLM_BudgetTable):
@@ -659,8 +689,9 @@ class KeyManagementSettings(LiteLLMBase):
 class TeamDefaultSettings(LiteLLMBase):
     team_id: str
 
-    class Config:
-        extra = "allow"  # allow params not defined here, these fall in litellm.completion(**kwargs)
+    model_config = ConfigDict(
+        extra = "allow",  # allow params not defined here, these fall in litellm.completion(**kwargs)
+    )
 
 
 class DynamoDBArgs(LiteLLMBase):
@@ -801,8 +832,9 @@ class ConfigYAML(LiteLLMBase):
         description="litellm router object settings. See router.py __init__ for all, example router.num_retries=5, router.timeout=5, router.max_retries=5, router.retry_after=5",
     )
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
 
 class LiteLLM_VerificationToken(LiteLLMBase):
@@ -836,8 +868,9 @@ class LiteLLM_VerificationToken(LiteLLMBase):
     user_id_rate_limits: Optional[dict] = None
     team_id_rate_limits: Optional[dict] = None
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
 
 class LiteLLM_VerificationTokenView(LiteLLM_VerificationToken):
@@ -867,7 +900,7 @@ class UserAPIKeyAuth(
     user_role: Optional[Literal["proxy_admin", "app_owner", "app_user"]] = None
     allowed_model_region: Optional[Literal["eu"]] = None
 
-    @root_validator(pre=True)
+    @model_validator(mode="before")
     def check_api_key(cls, values):
         if values.get("api_key") is not None:
             values.update({"token": hash_token(values.get("api_key"))})
@@ -894,7 +927,7 @@ class LiteLLM_UserTable(LiteLLMBase):
     tpm_limit: Optional[int] = None
     rpm_limit: Optional[int] = None
 
-    @root_validator(pre=True)
+    @model_validator(mode="before")
     def set_model_info(cls, values):
         if values.get("spend") is None:
             values.update({"spend": 0.0})
@@ -902,8 +935,9 @@ class LiteLLM_UserTable(LiteLLMBase):
             values.update({"models": []})
         return values
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
 
 class LiteLLM_EndUserTable(LiteLLMBase):
@@ -915,14 +949,15 @@ class LiteLLM_EndUserTable(LiteLLMBase):
     default_model: Optional[str] = None
     litellm_budget_table: Optional[LiteLLM_BudgetTable] = None
 
-    @root_validator(pre=True)
+    @model_validator(mode="before")
     def set_model_info(cls, values):
         if values.get("spend") is None:
             values.update({"spend": 0.0})
         return values
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
 
 class LiteLLM_SpendLogs(LiteLLMBase):

litellm/proxy/proxy_server.py

@@ -589,6 +589,15 @@ async def user_api_key_auth(
             )
 
             return _user_api_key_obj
+
+        ## IF it's not a master key
+        ## Route should not be in master_key_only_routes
+        if route in LiteLLMRoutes.master_key_only_routes.value:
+            raise Exception(
+                f"Tried to access route={route}, which is only for MASTER KEY"
+            )
+
+        ## Check DB
         if isinstance(
             api_key, str
         ):  # if generated token, make sure it starts with sk-.
@@ -3479,6 +3488,26 @@ async def startup_event():
             await proxy_config.add_deployment(
                 prisma_client=prisma_client, proxy_logging_obj=proxy_logging_obj
             )
+
+        if (
+            proxy_logging_obj is not None
+            and proxy_logging_obj.slack_alerting_instance is not None
+            and prisma_client is not None
+        ):
+            print("Alerting: Initializing Weekly/Monthly Spend Reports")  # noqa
+            ### Schedule weekly/monhtly spend reports ###
+            scheduler.add_job(
+                proxy_logging_obj.slack_alerting_instance.send_weekly_spend_report,
+                "cron",
+                day_of_week="mon",
+            )
+
+            scheduler.add_job(
+                proxy_logging_obj.slack_alerting_instance.send_monthly_spend_report,
+                "cron",
+                day=1,
+            )
+
         scheduler.start()
 
 
@@ -3698,8 +3727,9 @@ async def chat_completion(
         # skip router if user passed their key
         if "api_key" in data:
             tasks.append(litellm.acompletion(**data))
-        elif isinstance(data["model"], list) and llm_router is not None:
-            _models = data.pop("model")
+        elif "," in data["model"] and llm_router is not None:
+            _models_csv_string = data.pop("model")
+            _models = _models_csv_string.split(",")
             tasks.append(llm_router.abatch_completion(models=_models, **data))
         elif "user_config" in data:
             # initialize a new router instance. make request using this Router
@@ -3761,6 +3791,7 @@ async def chat_completion(
                 "x-litellm-cache-key": cache_key,
                 "x-litellm-model-api-base": api_base,
                 "x-litellm-version": version,
+                "x-litellm-model-region": user_api_key_dict.allowed_model_region or "",
             }
             selected_data_generator = select_data_generator(
                 response=response,
@@ -3777,6 +3808,9 @@ async def chat_completion(
         fastapi_response.headers["x-litellm-cache-key"] = cache_key
         fastapi_response.headers["x-litellm-model-api-base"] = api_base
         fastapi_response.headers["x-litellm-version"] = version
+        fastapi_response.headers["x-litellm-model-region"] = (
+            user_api_key_dict.allowed_model_region or ""
+        )
 
         ### CALL HOOKS ### - modify outgoing data
         response = await proxy_logging_obj.post_call_success_hook(
@@ -4161,6 +4195,9 @@ async def embeddings(
         fastapi_response.headers["x-litellm-cache-key"] = cache_key
         fastapi_response.headers["x-litellm-model-api-base"] = api_base
         fastapi_response.headers["x-litellm-version"] = version
+        fastapi_response.headers["x-litellm-model-region"] = (
+            user_api_key_dict.allowed_model_region or ""
+        )
 
         return response
     except Exception as e:
@@ -4330,6 +4367,9 @@ async def image_generation(
         fastapi_response.headers["x-litellm-cache-key"] = cache_key
         fastapi_response.headers["x-litellm-model-api-base"] = api_base
         fastapi_response.headers["x-litellm-version"] = version
+        fastapi_response.headers["x-litellm-model-region"] = (
+            user_api_key_dict.allowed_model_region or ""
+        )
 
         return response
     except Exception as e:
@@ -4523,6 +4563,9 @@ async def audio_transcriptions(
         fastapi_response.headers["x-litellm-cache-key"] = cache_key
         fastapi_response.headers["x-litellm-model-api-base"] = api_base
         fastapi_response.headers["x-litellm-version"] = version
+        fastapi_response.headers["x-litellm-model-region"] = (
+            user_api_key_dict.allowed_model_region or ""
+        )
 
         return response
     except Exception as e:
@@ -4698,6 +4741,9 @@ async def moderations(
         fastapi_response.headers["x-litellm-cache-key"] = cache_key
         fastapi_response.headers["x-litellm-model-api-base"] = api_base
         fastapi_response.headers["x-litellm-version"] = version
+        fastapi_response.headers["x-litellm-model-region"] = (
+            user_api_key_dict.allowed_model_region or ""
+        )
 
         return response
     except Exception as e:
@@ -5347,6 +5393,141 @@ async def view_spend_tags(
         )
 
 
+@router.get(
+    "/global/spend/report",
+    tags=["Budget & Spend Tracking"],
+    dependencies=[Depends(user_api_key_auth)],
+    include_in_schema=False,
+    responses={
+        200: {"model": List[LiteLLM_SpendLogs]},
+    },
+)
+async def get_global_spend_report(
+    start_date: Optional[str] = fastapi.Query(
+        default=None,
+        description="Time from which to start viewing spend",
+    ),
+    end_date: Optional[str] = fastapi.Query(
+        default=None,
+        description="Time till which to view spend",
+    ),
+):
+    """
+    Get Daily Spend per Team, based on specific startTime and endTime. Per team, view usage by each key, model
+    [
+        {
+            "group-by-day": "2024-05-10",
+            "teams": [
+                {
+                    "team_name": "team-1"
+                    "spend": 10,
+                    "keys": [
+                        "key": "1213",
+                        "usage": {
+                            "model-1": {
+                                    "cost": 12.50,
+                                    "input_tokens": 1000,
+                                    "output_tokens": 5000,
+                                    "requests": 100
+                                },
+                                "audio-modelname1": {
+                                "cost": 25.50,
+                                "seconds": 25,
+                                "requests": 50
+                        },
+                        }
+                    }
+            ]
+        ]
+    }
+    """
+    if start_date is None or end_date is None:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail={"error": "Please provide start_date and end_date"},
+        )
+
+    start_date_obj = datetime.strptime(start_date, "%Y-%m-%d")
+    end_date_obj = datetime.strptime(end_date, "%Y-%m-%d")
+
+    global prisma_client
+    try:
+        if prisma_client is None:
+            raise Exception(
+                f"Database not connected. Connect a database to your proxy - https://docs.litellm.ai/docs/simple_proxy#managing-auth---virtual-keys"
+            )
+
+        # first get data from spend logs -> SpendByModelApiKey
+        # then read data from "SpendByModelApiKey" to format the response obj
+        sql_query = """
+
+        WITH SpendByModelApiKey AS (
+            SELECT
+                date_trunc('day', sl."startTime") AS group_by_day,
+                COALESCE(tt.team_alias, 'Unassigned Team') AS team_name,
+                sl.model,
+                sl.api_key,
+                SUM(sl.spend) AS model_api_spend,
+                SUM(sl.total_tokens) AS model_api_tokens
+            FROM 
+                "LiteLLM_SpendLogs" sl
+            LEFT JOIN 
+                "LiteLLM_TeamTable" tt 
+            ON 
+                sl.team_id = tt.team_id
+            WHERE
+                sl."startTime" BETWEEN $1::date AND $2::date
+            GROUP BY
+                date_trunc('day', sl."startTime"),
+                tt.team_alias,
+                sl.model,
+                sl.api_key
+        )
+            SELECT
+                group_by_day,
+                jsonb_agg(jsonb_build_object(
+                    'team_name', team_name,
+                    'total_spend', total_spend,
+                    'metadata', metadata
+                )) AS teams
+            FROM (
+                SELECT
+                    group_by_day,
+                    team_name,
+                    SUM(model_api_spend) AS total_spend,
+                    jsonb_agg(jsonb_build_object(
+                        'model', model,
+                        'api_key', api_key,
+                        'spend', model_api_spend,
+                        'total_tokens', model_api_tokens
+                    )) AS metadata
+                FROM 
+                    SpendByModelApiKey
+                GROUP BY
+                    group_by_day,
+                    team_name
+            ) AS aggregated
+            GROUP BY
+                group_by_day
+            ORDER BY
+                group_by_day;
+            """
+
+        db_response = await prisma_client.db.query_raw(
+            sql_query, start_date_obj, end_date_obj
+        )
+        if db_response is None:
+            return []
+
+        return db_response
+
+    except Exception as e:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail={"error": str(e)},
+        )
+
+
 @router.get(
     "/global/spend/tags",
     tags=["Budget & Spend Tracking"],
@@ -5391,6 +5572,13 @@ async def global_view_spend_tags(
                 f"Database not connected. Connect a database to your proxy - https://docs.litellm.ai/docs/simple_proxy#managing-auth---virtual-keys"
             )
 
+        if end_date is None or start_date is None:
+            raise ProxyException(
+                message="Please provide start_date and end_date",
+                type="bad_request",
+                param=None,
+                code=status.HTTP_400_BAD_REQUEST,
+            )
         response = await ui_get_spend_by_tags(
             start_date=start_date, end_date=end_date, prisma_client=prisma_client
         )
@@ -5414,6 +5602,55 @@ async def global_view_spend_tags(
         )
 
 
+async def _get_spend_report_for_time_range(
+    start_date: str,
+    end_date: str,
+):
+    global prisma_client
+    if prisma_client is None:
+        verbose_proxy_logger.error(
+            f"Database not connected. Connect a database to your proxy for weekly, monthly spend reports"
+        )
+        return None
+
+    try:
+        sql_query = """
+        SELECT
+            t.team_alias,
+            SUM(s.spend) AS total_spend
+        FROM
+            "LiteLLM_SpendLogs" s
+        LEFT JOIN
+            "LiteLLM_TeamTable" t ON s.team_id = t.team_id
+        WHERE
+            s."startTime"::DATE >= $1::date AND s."startTime"::DATE <= $2::date
+        GROUP BY
+            t.team_alias
+        ORDER BY
+            total_spend DESC;
+        """
+        response = await prisma_client.db.query_raw(sql_query, start_date, end_date)
+
+        # get spend per tag for today
+        sql_query = """
+        SELECT 
+        jsonb_array_elements_text(request_tags) AS individual_request_tag,
+        SUM(spend) AS total_spend
+        FROM "LiteLLM_SpendLogs"
+        WHERE "startTime"::DATE >= $1::date AND "startTime"::DATE <= $2::date
+        GROUP BY individual_request_tag
+        ORDER BY total_spend DESC;
+        """
+
+        spend_per_tag = await prisma_client.db.query_raw(
+            sql_query, start_date, end_date
+        )
+
+        return response, spend_per_tag
+    except Exception as e:
+        verbose_proxy_logger.error("Exception in _get_daily_spend_reports", e)  # noqa
+
+
 @router.post(
     "/spend/calculate",
     tags=["Budget & Spend Tracking"],
@@ -5694,6 +5931,42 @@ async def view_spend_logs(
         )
 
 
+@router.post(
+    "/global/spend/reset",
+    tags=["Budget & Spend Tracking"],
+    dependencies=[Depends(user_api_key_auth)],
+)
+async def global_spend_reset():
+    """
+    ADMIN ONLY / MASTER KEY Only Endpoint
+
+    Globally reset spend for All API Keys and Teams, maintain LiteLLM_SpendLogs
+
+    1. LiteLLM_SpendLogs will maintain the logs on spend, no data gets deleted from there
+    2. LiteLLM_VerificationTokens spend will be set = 0
+    3. LiteLLM_TeamTable spend will be set = 0
+
+    """
+    global prisma_client
+    if prisma_client is None:
+        raise ProxyException(
+            message="Prisma Client is not initialized",
+            type="internal_error",
+            param="None",
+            code=status.HTTP_401_UNAUTHORIZED,
+        )
+
+    await prisma_client.db.litellm_verificationtoken.update_many(
+        data={"spend": 0.0}, where={}
+    )
+    await prisma_client.db.litellm_teamtable.update_many(data={"spend": 0.0}, where={})
+
+    return {
+        "message": "Spend for all API Keys and Teams reset successfully",
+        "status": "success",
+    }
+
+
 @router.get(
     "/global/spend/logs",
     tags=["Budget & Spend Tracking"],
@@ -5801,7 +6074,7 @@ async def global_spend_keys(
     tags=["Budget & Spend Tracking"],
     dependencies=[Depends(user_api_key_auth)],
 )
-async def global_spend_per_tea():
+async def global_spend_per_team():
     """
     [BETA] This is a beta endpoint. It will change.
 
@@ -9486,6 +9759,14 @@ async def health_services_endpoint(
                         level="Low",
                         alert_type="budget_alerts",
                     )
+
+                if prisma_client is not None:
+                    asyncio.create_task(
+                        proxy_logging_obj.slack_alerting_instance.send_monthly_spend_report()
+                    )
+                    asyncio.create_task(
+                        proxy_logging_obj.slack_alerting_instance.send_weekly_spend_report()
+                    )
                 return {
                     "status": "success",
                     "message": "Mock Slack Alert sent, verify Slack Alert Received on your channel",

litellm/router.py

@@ -9,7 +9,8 @@
 
 import copy, httpx
 from datetime import datetime
-from typing import Dict, List, Optional, Union, Literal, Any, BinaryIO, Tuple
+from typing import Dict, List, Optional, Union, Literal, Any, BinaryIO, Tuple, TypedDict
+from typing_extensions import overload
 import random, threading, time, traceback, uuid
 import litellm, openai, hashlib, json
 from litellm.caching import RedisCache, InMemoryCache, DualCache
@@ -46,6 +47,7 @@ from litellm.types.router import (
     updateLiteLLMParams,
     RetryPolicy,
     AlertingConfig,
+    DeploymentTypedDict,
 )
 from litellm.integrations.custom_logger import CustomLogger
 from litellm.llms.azure import get_azure_ad_token_from_oidc
@@ -61,7 +63,7 @@ class Router:
 
     def __init__(
         self,
-        model_list: Optional[list] = None,
+        model_list: Optional[List[Union[DeploymentTypedDict, Dict]]] = None,
         ## CACHING ##
         redis_url: Optional[str] = None,
         redis_host: Optional[str] = None,
@@ -82,6 +84,9 @@ class Router:
         default_max_parallel_requests: Optional[int] = None,
         set_verbose: bool = False,
         debug_level: Literal["DEBUG", "INFO"] = "INFO",
+        default_fallbacks: Optional[
+            List[str]
+        ] = None,  # generic fallbacks, works across all deployments
         fallbacks: List = [],
         context_window_fallbacks: List = [],
         model_group_alias: Optional[dict] = {},
@@ -258,6 +263,12 @@ class Router:
         self.retry_after = retry_after
         self.routing_strategy = routing_strategy
         self.fallbacks = fallbacks or litellm.fallbacks
+        if default_fallbacks is not None or litellm.default_fallbacks is not None:
+            _fallbacks = default_fallbacks or litellm.default_fallbacks
+            if self.fallbacks is not None:
+                self.fallbacks.append({"*": _fallbacks})
+            else:
+                self.fallbacks = [{"*": _fallbacks}]
         self.context_window_fallbacks = (
             context_window_fallbacks or litellm.context_window_fallbacks
         )
@@ -469,12 +480,30 @@ class Router:
             )
             raise e
 
+    # fmt: off
+
+    @overload
     async def acompletion(
-        self, model: str, messages: List[Dict[str, str]], **kwargs
-    ) -> Union[ModelResponse, CustomStreamWrapper]:
+        self, model: str, messages: List[Dict[str, str]], stream: Literal[True], **kwargs
+    ) -> CustomStreamWrapper: 
+        ...
+
+    @overload
+    async def acompletion(
+        self, model: str, messages: List[Dict[str, str]], stream: Literal[False] = False, **kwargs
+    ) -> ModelResponse: 
+        ...
+
+    # fmt: on
+
+    # The actual implementation of the function
+    async def acompletion(
+        self, model: str, messages: List[Dict[str, str]], stream=False, **kwargs
+    ):
         try:
             kwargs["model"] = model
             kwargs["messages"] = messages
+            kwargs["stream"] = stream
             kwargs["original_function"] = self._acompletion
             kwargs["num_retries"] = kwargs.get("num_retries", self.num_retries)
 
@@ -1413,7 +1442,7 @@ class Router:
                 verbose_router_logger.debug(f"Trying to fallback b/w models")
                 if (
                     hasattr(e, "status_code")
-                    and e.status_code == 400
+                    and e.status_code == 400  # type: ignore
                     and not isinstance(e, litellm.ContextWindowExceededError)
                 ):  # don't retry a malformed request
                     raise e
@@ -1444,18 +1473,29 @@ class Router:
                             response = await self.async_function_with_retries(
                                 *args, **kwargs
                             )
+                            verbose_router_logger.info(
+                                "Successful fallback b/w models."
+                            )
                             return response
                         except Exception as e:
                             pass
                 elif fallbacks is not None:
                     verbose_router_logger.debug(f"inside model fallbacks: {fallbacks}")
-                    for item in fallbacks:
-                        key_list = list(item.keys())
-                        if len(key_list) == 0:
-                            continue
-                        if key_list[0] == model_group:
+                    generic_fallback_idx: Optional[int] = None
+                    ## check for specific model group-specific fallbacks
+                    for idx, item in enumerate(fallbacks):
+                        if list(item.keys())[0] == model_group:
                             fallback_model_group = item[model_group]
                             break
+                        elif list(item.keys())[0] == "*":
+                            generic_fallback_idx = idx
+                    ## if none, check for generic fallback
+                    if (
+                        fallback_model_group is None
+                        and generic_fallback_idx is not None
+                    ):
+                        fallback_model_group = fallbacks[generic_fallback_idx]["*"]
+
                     if fallback_model_group is None:
                         verbose_router_logger.info(
                             f"No fallback model group found for original model_group={model_group}. Fallbacks={fallbacks}"
@@ -1478,6 +1518,9 @@ class Router:
                             response = await self.async_function_with_fallbacks(
                                 *args, **kwargs
                             )
+                            verbose_router_logger.info(
+                                "Successful fallback b/w models."
+                            )
                             return response
                         except Exception as e:
                             raise e
@@ -1512,7 +1555,7 @@ class Router:
              
             """
             _healthy_deployments = await self._async_get_healthy_deployments(
-                model=kwargs.get("model"),
+                model=kwargs.get("model") or "",
             )
 
             # raises an exception if this error should not be retries
@@ -1619,12 +1662,18 @@ class Router:
         Try calling the function_with_retries
         If it fails after num_retries, fall back to another model group
         """
+        mock_testing_fallbacks = kwargs.pop("mock_testing_fallbacks", None)
         model_group = kwargs.get("model")
         fallbacks = kwargs.get("fallbacks", self.fallbacks)
         context_window_fallbacks = kwargs.get(
             "context_window_fallbacks", self.context_window_fallbacks
         )
         try:
+            if mock_testing_fallbacks is not None and mock_testing_fallbacks == True:
+                raise Exception(
+                    f"This is a mock exception for model={model_group}, to trigger a fallback. Fallbacks={fallbacks}"
+                )
+
             response = self.function_with_retries(*args, **kwargs)
             return response
         except Exception as e:
@@ -1633,7 +1682,7 @@ class Router:
             try:
                 if (
                     hasattr(e, "status_code")
-                    and e.status_code == 400
+                    and e.status_code == 400  # type: ignore
                     and not isinstance(e, litellm.ContextWindowExceededError)
                 ):  # don't retry a malformed request
                     raise e
@@ -1675,10 +1724,20 @@ class Router:
                 elif fallbacks is not None:
                     verbose_router_logger.debug(f"inside model fallbacks: {fallbacks}")
                     fallback_model_group = None
-                    for item in fallbacks:
+                    generic_fallback_idx: Optional[int] = None
+                    ## check for specific model group-specific fallbacks
+                    for idx, item in enumerate(fallbacks):
                         if list(item.keys())[0] == model_group:
                             fallback_model_group = item[model_group]
                             break
+                        elif list(item.keys())[0] == "*":
+                            generic_fallback_idx = idx
+                    ## if none, check for generic fallback
+                    if (
+                        fallback_model_group is None
+                        and generic_fallback_idx is not None
+                    ):
+                        fallback_model_group = fallbacks[generic_fallback_idx]["*"]
 
                     if fallback_model_group is None:
                         raise original_exception
@@ -3259,13 +3318,12 @@ class Router:
             healthy_deployments.remove(deployment)
 
         # filter pre-call checks
+        _allowed_model_region = (
+            request_kwargs.get("allowed_model_region")
+            if request_kwargs is not None
+            else None
+        )
         if self.enable_pre_call_checks and messages is not None:
-            _allowed_model_region = (
-                request_kwargs.get("allowed_model_region")
-                if request_kwargs is not None
-                else None
-            )
-
             if _allowed_model_region == "eu":
                 healthy_deployments = self._pre_call_checks(
                     model=model,
@@ -3286,8 +3344,10 @@ class Router:
                 )
 
         if len(healthy_deployments) == 0:
+            if _allowed_model_region is None:
+                _allowed_model_region = "n/a"
             raise ValueError(
-                f"{RouterErrors.no_deployments_available.value}, passed model={model}"
+                f"{RouterErrors.no_deployments_available.value}, passed model={model}. Enable pre-call-checks={self.enable_pre_call_checks}, allowed_model_region={_allowed_model_region}"
             )
 
         if (
@@ -3647,7 +3707,7 @@ class Router:
                 )
                 asyncio.create_task(
                     proxy_logging_obj.slack_alerting_instance.send_alert(
-                        message=f"Router: Cooling down deployment: {_api_base}, for {self.cooldown_time} seconds. Got exception: {str(exception_status)}",
+                        message=f"Router: Cooling down deployment: {_api_base}, for {self.cooldown_time} seconds. Got exception: {str(exception_status)}. Change 'cooldown_time' + 'allowed_fails' under 'Router Settings' on proxy UI, or via config - https://docs.litellm.ai/docs/proxy/reliability#fallbacks--retries--timeouts--cooldowns",
                         alert_type="cooldown_deployment",
                         level="Low",
                     )

litellm/router_strategy/lowest_cost.py

@@ -101,9 +101,6 @@ class LowestCostLoggingHandler(CustomLogger):
                 if precise_minute not in request_count_dict[id]:
                     request_count_dict[id][precise_minute] = {}
 
-                if precise_minute not in request_count_dict[id]:
-                    request_count_dict[id][precise_minute] = {}
-
                 ## TPM
                 request_count_dict[id][precise_minute]["tpm"] = (
                     request_count_dict[id][precise_minute].get("tpm", 0) + total_tokens

litellm/router_strategy/lowest_latency.py

@@ -115,9 +115,6 @@ class LowestLatencyLoggingHandler(CustomLogger):
                 if precise_minute not in request_count_dict[id]:
                     request_count_dict[id][precise_minute] = {}
 
-                if precise_minute not in request_count_dict[id]:
-                    request_count_dict[id][precise_minute] = {}
-
                 ## TPM
                 request_count_dict[id][precise_minute]["tpm"] = (
                     request_count_dict[id][precise_minute].get("tpm", 0) + total_tokens

litellm/tests/test_alangfuse.py

@@ -228,6 +228,40 @@ async def test_langfuse_logging_without_request_response(stream, langfuse_client
         pytest.fail(f"An exception occurred - {e}")
 
 
+@pytest.mark.asyncio
+async def test_langfuse_masked_input_output(langfuse_client):
+    """
+    Test that creates a trace with masked input and output
+    """
+    import uuid
+
+    for mask_value in [True, False]:
+        _unique_trace_name = f"litellm-test-{str(uuid.uuid4())}"
+        litellm.set_verbose = True
+        litellm.success_callback = ["langfuse"]
+        response = await create_async_task(
+            model="gpt-3.5-turbo",
+            messages=[{"role": "user", "content": "This is a test"}],
+            metadata={"trace_id": _unique_trace_name, "mask_input": mask_value, "mask_output": mask_value},
+            mock_response="This is a test response"
+        )
+        print(response)
+        expected_input = "redacted-by-litellm" if mask_value else {'messages': [{'content': 'This is a test', 'role': 'user'}]}
+        expected_output = "redacted-by-litellm" if mask_value else {'content': 'This is a test response', 'role': 'assistant'}
+        langfuse_client.flush()
+        await asyncio.sleep(2)
+
+        # get trace with _unique_trace_name
+        trace = langfuse_client.get_trace(id=_unique_trace_name)
+        generations = list(
+            reversed(langfuse_client.get_generations(trace_id=_unique_trace_name).data)
+        )
+
+        assert trace.input == expected_input
+        assert trace.output == expected_output
+        assert generations[0].input == expected_input
+        assert generations[0].output == expected_output
+
 @pytest.mark.asyncio
 async def test_langfuse_logging_metadata(langfuse_client):
     """

litellm/tests/test_alerting.py

@@ -359,3 +359,49 @@ async def test_send_llm_exception_to_slack():
     )
 
     await asyncio.sleep(3)
+
+
+# test models with 0 metrics are ignored
+@pytest.mark.asyncio
+async def test_send_daily_reports_ignores_zero_values():
+    router = MagicMock()
+    router.get_model_ids.return_value = ['model1', 'model2', 'model3']
+    
+    slack_alerting = SlackAlerting(internal_usage_cache=MagicMock())
+    # model1:failed=None, model2:failed=0, model3:failed=10, model1:latency=0; model2:latency=0; model3:latency=None
+    slack_alerting.internal_usage_cache.async_batch_get_cache = AsyncMock(return_value=[None, 0, 10, 0, 0, None])
+    slack_alerting.internal_usage_cache.async_batch_set_cache = AsyncMock()
+
+    router.get_model_info.side_effect = lambda x: {"litellm_params": {"model": x}}
+    
+    with patch.object(slack_alerting, 'send_alert', new=AsyncMock()) as mock_send_alert:
+        result = await slack_alerting.send_daily_reports(router)
+        
+        # Check that the send_alert method was called
+        mock_send_alert.assert_called_once()
+        message = mock_send_alert.call_args[1]['message']
+        
+        # Ensure the message includes only the non-zero, non-None metrics
+        assert "model3" in message
+        assert "model2" not in message
+        assert "model1" not in message
+    
+    assert result == True
+
+
+# test no alert is sent if all None or 0 metrics
+@pytest.mark.asyncio
+async def test_send_daily_reports_all_zero_or_none():
+    router = MagicMock()
+    router.get_model_ids.return_value = ['model1', 'model2', 'model3']
+    
+    slack_alerting = SlackAlerting(internal_usage_cache=MagicMock())
+    slack_alerting.internal_usage_cache.async_batch_get_cache = AsyncMock(return_value=[None, 0, None, 0, None, 0])
+    
+    with patch.object(slack_alerting, 'send_alert', new=AsyncMock()) as mock_send_alert:
+        result = await slack_alerting.send_daily_reports(router)
+        
+        # Check that the send_alert method was not called
+        mock_send_alert.assert_not_called()
+    
+    assert result == False

litellm/tests/test_amazing_vertex_completion.py

@@ -590,19 +590,20 @@ def test_gemini_pro_vision_base64():
             pytest.fail(f"An exception occurred - {str(e)}")
 
 
+@pytest.mark.parametrize("sync_mode", [True, False])
 @pytest.mark.asyncio
-def test_gemini_pro_function_calling():
+async def test_gemini_pro_function_calling(sync_mode):
     try:
         load_vertex_ai_credentials()
-        response = litellm.completion(
-            model="vertex_ai/gemini-pro",
-            messages=[
+        data = {
+            "model": "vertex_ai/gemini-pro",
+            "messages": [
                 {
                     "role": "user",
                     "content": "Call the submit_cities function with San Francisco and New York",
                 }
             ],
-            tools=[
+            "tools": [
                 {
                     "type": "function",
                     "function": {
@@ -618,11 +619,13 @@ def test_gemini_pro_function_calling():
                     },
                 }
             ],
-        )
+        }
+        if sync_mode:
+            response = litellm.completion(**data)
+        else:
+            response = await litellm.acompletion(**data)
 
         print(f"response: {response}")
-    except litellm.APIError as e:
-        pass
     except litellm.RateLimitError as e:
         pass
     except Exception as e:
@@ -635,73 +638,66 @@ def test_gemini_pro_function_calling():
 # gemini_pro_function_calling()
 
 
-@pytest.mark.parametrize("stream", [False, True])
 @pytest.mark.parametrize("sync_mode", [False, True])
 @pytest.mark.asyncio
-async def test_gemini_pro_function_calling_streaming(stream, sync_mode):
+async def test_gemini_pro_function_calling_streaming(sync_mode):
     load_vertex_ai_credentials()
     litellm.set_verbose = True
-    tools = [
-        {
-            "type": "function",
-            "function": {
-                "name": "get_current_weather",
-                "description": "Get the current weather in a given location",
-                "parameters": {
-                    "type": "object",
-                    "properties": {
-                        "location": {
-                            "type": "string",
-                            "description": "The city and state, e.g. San Francisco, CA",
+    data = {
+        "model": "vertex_ai/gemini-pro",
+        "messages": [
+            {
+                "role": "user",
+                "content": "Call the submit_cities function with San Francisco and New York",
+            }
+        ],
+        "tools": [
+            {
+                "type": "function",
+                "function": {
+                    "name": "submit_cities",
+                    "description": "Submits a list of cities",
+                    "parameters": {
+                        "type": "object",
+                        "properties": {
+                            "cities": {"type": "array", "items": {"type": "string"}}
                         },
-                        "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]},
+                        "required": ["cities"],
                     },
-                    "required": ["location"],
                 },
-            },
-        }
-    ]
-    messages = [
-        {
-            "role": "user",
-            "content": "What's the weather like in Boston today in fahrenheit?",
-        }
-    ]
-    optional_params = {
-        "tools": tools,
+            }
+        ],
         "tool_choice": "auto",
         "n": 1,
-        "stream": stream,
+        "stream": True,
         "temperature": 0.1,
     }
+    chunks = []
     try:
         if sync_mode == True:
-            response = litellm.completion(
-                model="gemini-pro", messages=messages, **optional_params
-            )
+            response = litellm.completion(**data)
             print(f"completion: {response}")
 
-            if stream == True:
-                # assert completion.choices[0].message.content is None
-                # assert len(completion.choices[0].message.tool_calls) == 1
-                for chunk in response:
-                    assert isinstance(chunk, litellm.ModelResponse)
-            else:
-                assert isinstance(response, litellm.ModelResponse)
+            for chunk in response:
+                chunks.append(chunk)
+                assert isinstance(chunk, litellm.ModelResponse)
         else:
-            response = await litellm.acompletion(
-                model="gemini-pro", messages=messages, **optional_params
-            )
+            response = await litellm.acompletion(**data)
             print(f"completion: {response}")
 
-            if stream == True:
-                # assert completion.choices[0].message.content is None
-                # assert len(completion.choices[0].message.tool_calls) == 1
-                async for chunk in response:
-                    print(f"chunk: {chunk}")
-                    assert isinstance(chunk, litellm.ModelResponse)
-            else:
-                assert isinstance(response, litellm.ModelResponse)
+            assert isinstance(response, litellm.CustomStreamWrapper)
+
+            async for chunk in response:
+                print(f"chunk: {chunk}")
+                chunks.append(chunk)
+                assert isinstance(chunk, litellm.ModelResponse)
+
+        complete_response = litellm.stream_chunk_builder(chunks=chunks)
+        assert (
+            complete_response.choices[0].message.content is not None
+            or len(complete_response.choices[0].message.tool_calls) > 0
+        )
+        print(f"complete_response: {complete_response}")
     except litellm.APIError as e:
         pass
     except litellm.RateLimitError as e:

litellm/tests/test_azure_perf.py

@@ -26,7 +26,7 @@ model_list = [
     }
 ]
 
-router = litellm.Router(model_list=model_list)
+router = litellm.Router(model_list=model_list)  # type: ignore
 
 
 async def _openai_completion():

litellm/tests/test_caching.py

@@ -599,7 +599,10 @@ def test_redis_cache_completion():
     )
     print("test2 for Redis Caching - non streaming")
     response1 = completion(
-        model="gpt-3.5-turbo", messages=messages, caching=True, max_tokens=20
+        model="gpt-3.5-turbo",
+        messages=messages,
+        caching=True,
+        max_tokens=20,
     )
     response2 = completion(
         model="gpt-3.5-turbo", messages=messages, caching=True, max_tokens=20
@@ -653,7 +656,6 @@ def test_redis_cache_completion():
     assert response1.created == response2.created
     assert response1.choices[0].message.content == response2.choices[0].message.content
 
-
 # test_redis_cache_completion()
 
 
@@ -875,6 +877,80 @@ async def test_redis_cache_acompletion_stream_bedrock():
         print(e)
         raise e
 
+def test_disk_cache_completion():
+    litellm.set_verbose = False
+
+    random_number = random.randint(
+        1, 100000
+    )  # add a random number to ensure it's always adding / reading from cache
+    messages = [
+        {"role": "user", "content": f"write a one sentence poem about: {random_number}"}
+    ]
+    litellm.cache = Cache(
+        type="disk",
+    )
+
+    response1 = completion(
+        model="gpt-3.5-turbo",
+        messages=messages,
+        caching=True,
+        max_tokens=20,
+        mock_response="This number is so great!",
+    )
+    # response2 is mocked to a different response from response1,
+    # but the completion from the cache should be used instead of the mock
+    # response since the input is the same as response1
+    response2 = completion(
+        model="gpt-3.5-turbo",
+        messages=messages,
+        caching=True,
+        max_tokens=20,
+        mock_response="This number is awful!",
+    )
+    # Since the parameters are not the same as response1, response3 should actually
+    # be the mock response
+    response3 = completion(
+        model="gpt-3.5-turbo",
+        messages=messages,
+        caching=True,
+        temperature=0.5,
+        mock_response="This number is awful!",
+    )
+
+    print("\nresponse 1", response1)
+    print("\nresponse 2", response2)
+    print("\nresponse 3", response3)
+    # print("\nresponse 4", response4)
+    litellm.cache = None
+    litellm.success_callback = []
+    litellm._async_success_callback = []
+
+    # 1 & 2 should be exactly the same 
+    # 1 & 3 should be different, since input params are diff
+    if (
+        response1["choices"][0]["message"]["content"]
+        != response2["choices"][0]["message"]["content"]
+    ):  # 1 and 2 should be the same
+        # 1&2 have the exact same input params. This MUST Be a CACHE HIT
+        print(f"response1: {response1}")
+        print(f"response2: {response2}")
+        pytest.fail(f"Error occurred:")
+    if (
+        response1["choices"][0]["message"]["content"]
+        == response3["choices"][0]["message"]["content"]
+    ):
+        # if input params like max_tokens, temperature are diff it should NOT be a cache hit
+        print(f"response1: {response1}")
+        print(f"response3: {response3}")
+        pytest.fail(
+            f"Response 1 == response 3. Same model, diff params shoudl not cache Error"
+            f" occurred:"
+        )
+
+    assert response1.id == response2.id
+    assert response1.created == response2.created
+    assert response1.choices[0].message.content == response2.choices[0].message.content
+
 
 @pytest.mark.skip(reason="AWS Suspended Account")
 @pytest.mark.asyncio

litellm/tests/test_completion.py

@@ -68,6 +68,51 @@ def test_completion_custom_provider_model_name():
         pytest.fail(f"Error occurred: {e}")
 
 
+def _openai_mock_response(*args, **kwargs) -> litellm.ModelResponse:
+    _data = {
+        "id": "chatcmpl-123",
+        "object": "chat.completion",
+        "created": 1677652288,
+        "model": "gpt-3.5-turbo-0125",
+        "system_fingerprint": "fp_44709d6fcb",
+        "choices": [
+            {
+                "index": 0,
+                "message": {
+                    "role": None,
+                    "content": "\n\nHello there, how may I assist you today?",
+                },
+                "logprobs": None,
+                "finish_reason": "stop",
+            }
+        ],
+        "usage": {"prompt_tokens": 9, "completion_tokens": 12, "total_tokens": 21},
+    }
+    return litellm.ModelResponse(**_data)
+
+
+def test_null_role_response():
+    """
+    Test if api returns 'null' role, 'assistant' role is still returned
+    """
+    import openai
+
+    openai_client = openai.OpenAI()
+    with patch.object(
+        openai_client.chat.completions, "create", side_effect=_openai_mock_response
+    ) as mock_response:
+        response = litellm.completion(
+            model="gpt-3.5-turbo",
+            messages=[{"role": "user", "content": "Hey! how's it going?"}],
+            client=openai_client,
+        )
+        print(f"response: {response}")
+
+        assert response.id == "chatcmpl-123"
+
+        assert response.choices[0].message.role == "assistant"
+
+
 def test_completion_azure_command_r():
     try:
         litellm.set_verbose = True
@@ -665,6 +710,7 @@ def test_completion_mistral_api():
                     "content": "Hey, how's it going?",
                 }
             ],
+            seed=10,
         )
         # Add any assertions here to check the response
         print(response)
@@ -839,7 +885,7 @@ async def test_acompletion_claude2_1():
             },
             {"role": "user", "content": "Generate a 3 liner joke for me"},
         ]
-        # test without max tokens
+        # test without max-tokens
         response = await litellm.acompletion(model="claude-2.1", messages=messages)
         # Add any assertions here to check the response
         print(response)
@@ -3296,6 +3342,8 @@ def test_completion_watsonx():
         print(response)
     except litellm.APIError as e:
         pass
+    except litellm.RateLimitError as e:
+        pass
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
 
@@ -3315,6 +3363,8 @@ def test_completion_stream_watsonx():
             print(chunk)
     except litellm.APIError as e:
         pass
+    except litellm.RateLimitError as e:
+        pass
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
 
@@ -3379,6 +3429,8 @@ async def test_acompletion_watsonx():
         )
         # Add any assertions here to check the response
         print(response)
+    except litellm.RateLimitError as e:
+        pass
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
 
@@ -3399,6 +3451,8 @@ async def test_acompletion_stream_watsonx():
         # Add any assertions here to check the response
         async for chunk in response:
             print(chunk)
+    except litellm.RateLimitError as e:
+        pass
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
 

litellm/tests/test_completion_cost.py

@@ -5,6 +5,7 @@ sys.path.insert(
     0, os.path.abspath("../..")
 )  # Adds the parent directory to the system path
 import time
+from typing import Optional
 import litellm
 from litellm import (
     get_max_tokens,
@@ -12,7 +13,56 @@ from litellm import (
     open_ai_chat_completion_models,
     TranscriptionResponse,
 )
-import pytest
+from litellm.utils import CustomLogger
+import pytest, asyncio
+
+
+class CustomLoggingHandler(CustomLogger):
+    response_cost: Optional[float] = None
+
+    def __init__(self):
+        super().__init__()
+
+    def log_success_event(self, kwargs, response_obj, start_time, end_time):
+        self.response_cost = kwargs["response_cost"]
+
+    async def async_log_success_event(self, kwargs, response_obj, start_time, end_time):
+        print(f"kwargs - {kwargs}")
+        print(f"kwargs response cost - {kwargs.get('response_cost')}")
+        self.response_cost = kwargs["response_cost"]
+
+        print(f"response_cost: {self.response_cost} ")
+
+
+@pytest.mark.parametrize("sync_mode", [True, False])
+@pytest.mark.asyncio
+async def test_custom_pricing(sync_mode):
+    new_handler = CustomLoggingHandler()
+    litellm.callbacks = [new_handler]
+    if sync_mode:
+        response = litellm.completion(
+            model="gpt-3.5-turbo",
+            messages=[{"role": "user", "content": "Hey!"}],
+            mock_response="What do you want?",
+            input_cost_per_token=0.0,
+            output_cost_per_token=0.0,
+        )
+        time.sleep(5)
+    else:
+        response = await litellm.acompletion(
+            model="gpt-3.5-turbo",
+            messages=[{"role": "user", "content": "Hey!"}],
+            mock_response="What do you want?",
+            input_cost_per_token=0.0,
+            output_cost_per_token=0.0,
+        )
+
+        await asyncio.sleep(5)
+
+    print(f"new_handler.response_cost: {new_handler.response_cost}")
+    assert new_handler.response_cost is not None
+
+    assert new_handler.response_cost == 0
 
 
 def test_get_gpt3_tokens():

litellm/tests/test_config.py

@@ -5,6 +5,7 @@
 import sys, os
 import traceback
 from dotenv import load_dotenv
+from pydantic import ConfigDict
 
 load_dotenv()
 import os, io
@@ -25,9 +26,7 @@ class DBModel(BaseModel):
     model_name: str
     model_info: dict
     litellm_params: dict
-
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(protected_namespaces=())
 
 
 @pytest.mark.asyncio

litellm/tests/test_embedding.py

@@ -494,6 +494,8 @@ def test_watsonx_embeddings():
         )
         print(f"response: {response}")
         assert isinstance(response.usage, litellm.Usage)
+    except litellm.RateLimitError as e:
+        pass
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
 

litellm/tests/test_function_calling.py

@@ -37,14 +37,19 @@ def get_current_weather(location, unit="fahrenheit"):
 
 
 # Example dummy function hard coded to return the same weather
+
+
 # In production, this could be your backend API or an external API
-def test_parallel_function_call():
+@pytest.mark.parametrize(
+    "model", ["gpt-3.5-turbo-1106", "mistral/mistral-large-latest"]
+)
+def test_parallel_function_call(model):
     try:
         # Step 1: send the conversation and available functions to the model
         messages = [
             {
                 "role": "user",
-                "content": "What's the weather like in San Francisco, Tokyo, and Paris?",
+                "content": "What's the weather like in San Francisco, Tokyo, and Paris? - give me 3 responses",
             }
         ]
         tools = [
@@ -58,7 +63,7 @@ def test_parallel_function_call():
                         "properties": {
                             "location": {
                                 "type": "string",
-                                "description": "The city and state, e.g. San Francisco, CA",
+                                "description": "The city and state",
                             },
                             "unit": {
                                 "type": "string",
@@ -71,7 +76,7 @@ def test_parallel_function_call():
             }
         ]
         response = litellm.completion(
-            model="gpt-3.5-turbo-1106",
+            model=model,
             messages=messages,
             tools=tools,
             tool_choice="auto",  # auto is default, but we'll be explicit
@@ -83,8 +88,8 @@ def test_parallel_function_call():
         print("length of tool calls", len(tool_calls))
         print("Expecting there to be 3 tool calls")
         assert (
-            len(tool_calls) > 1
-        )  # this has to call the function for SF, Tokyo and parise
+            len(tool_calls) > 0
+        )  # this has to call the function for SF, Tokyo and paris
 
         # Step 2: check if the model wanted to call a function
         if tool_calls:
@@ -116,7 +121,7 @@ def test_parallel_function_call():
                 )  # extend conversation with function response
             print(f"messages: {messages}")
             second_response = litellm.completion(
-                model="gpt-3.5-turbo-1106", messages=messages, temperature=0.2, seed=22
+                model=model, messages=messages, temperature=0.2, seed=22
             )  # get a new response from the model where it can see the function response
             print("second response\n", second_response)
             return second_response

litellm/tests/test_key_generate_prisma.py

@@ -2013,3 +2013,74 @@ async def test_master_key_hashing(prisma_client):
     except Exception as e:
         print("Got Exception", e)
         pytest.fail(f"Got exception {e}")
+
+
+@pytest.mark.asyncio
+async def test_reset_spend_authentication(prisma_client):
+    """
+    1. Test master key can access this route  -> ONLY MASTER KEY SHOULD BE ABLE TO RESET SPEND
+    2. Test that non-master key gets rejected
+    3. Test that non-master key with role == "proxy_admin" or admin gets rejected
+    """
+
+    print("prisma client=", prisma_client)
+
+    master_key = "sk-1234"
+
+    setattr(litellm.proxy.proxy_server, "prisma_client", prisma_client)
+    setattr(litellm.proxy.proxy_server, "master_key", master_key)
+
+    await litellm.proxy.proxy_server.prisma_client.connect()
+    from litellm.proxy.proxy_server import user_api_key_cache
+
+    bearer_token = "Bearer " + master_key
+
+    request = Request(scope={"type": "http"})
+    request._url = URL(url="/global/spend/reset")
+
+    # Test 1 - Master Key
+    result: UserAPIKeyAuth = await user_api_key_auth(
+        request=request, api_key=bearer_token
+    )
+
+    print("result from user auth with Master key", result)
+    assert result.token is not None
+
+    # Test 2 - Non-Master Key
+    _response = await new_user(
+        data=NewUserRequest(
+            tpm_limit=20,
+        )
+    )
+
+    generate_key = "Bearer " + _response.key
+
+    try:
+        await user_api_key_auth(request=request, api_key=generate_key)
+        pytest.fail(f"This should have failed!. IT's an expired key")
+    except Exception as e:
+        print("Got Exception", e)
+        assert (
+            "Tried to access route=/global/spend/reset, which is only for MASTER KEY"
+            in e.message
+        )
+
+    # Test 3 - Non-Master Key with role == "proxy_admin" or admin
+    _response = await new_user(
+        data=NewUserRequest(
+            user_role="proxy_admin",
+            tpm_limit=20,
+        )
+    )
+
+    generate_key = "Bearer " + _response.key
+
+    try:
+        await user_api_key_auth(request=request, api_key=generate_key)
+        pytest.fail(f"This should have failed!. IT's an expired key")
+    except Exception as e:
+        print("Got Exception", e)
+        assert (
+            "Tried to access route=/global/spend/reset, which is only for MASTER KEY"
+            in e.message
+        )

litellm/tests/test_logfire.py

@@ -0,0 +1,117 @@
+import sys
+import os
+import json
+import time
+
+import logfire
+import litellm
+import pytest
+from logfire.testing import TestExporter, SimpleSpanProcessor
+
+sys.path.insert(0, os.path.abspath("../.."))
+
+# Testing scenarios for logfire logging:
+# 1. Test logfire logging for completion
+# 2. Test logfire logging for acompletion
+# 3. Test logfire logging for completion while streaming is enabled
+# 4. Test logfire logging for completion while streaming is enabled
+
+
+@pytest.mark.parametrize("stream", [False, True])
+def test_completion_logfire_logging(stream):
+    litellm.success_callback = ["logfire"]
+    litellm.set_verbose = True
+
+    exporter = TestExporter()
+    logfire.configure(
+        send_to_logfire=False,
+        console=False,
+        processors=[SimpleSpanProcessor(exporter)],
+        collect_system_metrics=False,
+    )
+    messages = [{"role": "user", "content": "what llm are u"}]
+    temperature = 0.3
+    max_tokens = 10
+    response = litellm.completion(
+        model="gpt-3.5-turbo",
+        messages=messages,
+        max_tokens=max_tokens,
+        temperature=temperature,
+        stream=stream,
+    )
+    print(response)
+
+    if stream:
+        for chunk in response:
+            print(chunk)
+
+    time.sleep(5)
+    exported_spans = exporter.exported_spans_as_dict()
+
+    assert len(exported_spans) == 1
+    assert (
+        exported_spans[0]["attributes"]["logfire.msg"]
+        == "Chat Completion with 'gpt-3.5-turbo'"
+    )
+
+    request_data = json.loads(exported_spans[0]["attributes"]["request_data"])
+
+    assert request_data["model"] == "gpt-3.5-turbo"
+    assert request_data["messages"] == messages
+
+    assert "completion_tokens" in request_data["usage"]
+    assert "prompt_tokens" in request_data["usage"]
+    assert "total_tokens" in request_data["usage"]
+    assert request_data["response"]["choices"][0]["message"]["content"]
+    assert request_data["modelParameters"]["max_tokens"] == max_tokens
+    assert request_data["modelParameters"]["temperature"] == temperature
+
+
+@pytest.mark.asyncio
+@pytest.mark.parametrize("stream", [False, True])
+async def test_acompletion_logfire_logging(stream):
+    litellm.success_callback = ["logfire"]
+    litellm.set_verbose = True
+
+    exporter = TestExporter()
+    logfire.configure(
+        send_to_logfire=False,
+        console=False,
+        processors=[SimpleSpanProcessor(exporter)],
+        collect_system_metrics=False,
+    )
+    messages = [{"role": "user", "content": "what llm are u"}]
+    temperature = 0.3
+    max_tokens = 10
+    response = await litellm.acompletion(
+        model="gpt-3.5-turbo",
+        messages=messages,
+        max_tokens=max_tokens,
+        temperature=temperature,
+    )
+    print(response)
+    if stream:
+        for chunk in response:
+            print(chunk)
+
+    time.sleep(5)
+    exported_spans = exporter.exported_spans_as_dict()
+    print("exported_spans", exported_spans)
+
+    assert len(exported_spans) == 1
+    assert (
+        exported_spans[0]["attributes"]["logfire.msg"]
+        == "Chat Completion with 'gpt-3.5-turbo'"
+    )
+
+    request_data = json.loads(exported_spans[0]["attributes"]["request_data"])
+
+    assert request_data["model"] == "gpt-3.5-turbo"
+    assert request_data["messages"] == messages
+
+    assert "completion_tokens" in request_data["usage"]
+    assert "prompt_tokens" in request_data["usage"]
+    assert "total_tokens" in request_data["usage"]
+    assert request_data["response"]["choices"][0]["message"]["content"]
+    assert request_data["modelParameters"]["max_tokens"] == max_tokens
+    assert request_data["modelParameters"]["temperature"] == temperature

litellm/tests/test_router.py

@@ -134,11 +134,13 @@ async def test_router_retries(sync_mode):
             messages=[{"role": "user", "content": "Hey, how's it going?"}],
         )
     else:
-        await router.acompletion(
+        response = await router.acompletion(
             model="gpt-3.5-turbo",
             messages=[{"role": "user", "content": "Hey, how's it going?"}],
         )
 
+        print(response.choices[0].message)
+
 
 @pytest.mark.parametrize(
     "mistral_api_base",

litellm/tests/test_router_debug_logs.py

@@ -85,6 +85,7 @@ def test_async_fallbacks(caplog):
         "litellm.acompletion(model=gpt-3.5-turbo)\x1b[31m Exception OpenAIException - Error code: 401 - {'error': {'message': 'Incorrect API key provided: bad-key. You can find your API key at https://platform.openai.com/account/api-keys.', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_api_key'}} \nModel: gpt-3.5-turbo\nAPI Base: https://api.openai.com\nMessages: [{'content': 'Hello, how are you?', 'role': 'user'}]\nmodel_group: gpt-3.5-turbo\n\ndeployment: gpt-3.5-turbo\n\x1b[0m",
         "Falling back to model_group = azure/gpt-3.5-turbo",
         "litellm.acompletion(model=azure/chatgpt-v-2)\x1b[32m 200 OK\x1b[0m",
+        "Successful fallback b/w models.",
     ]
 
     # Assert that the captured logs match the expected log messages

litellm/tests/test_router_fallbacks.py

@@ -961,3 +961,101 @@ def test_custom_cooldown_times():
 
     except Exception as e:
         print(e)
+
+
+@pytest.mark.parametrize("sync_mode", [True, False])
+@pytest.mark.asyncio
+async def test_service_unavailable_fallbacks(sync_mode):
+    """
+    Initial model - openai
+    Fallback - azure
+
+    Error - 503, service unavailable
+    """
+    router = Router(
+        model_list=[
+            {
+                "model_name": "gpt-3.5-turbo-012",
+                "litellm_params": {
+                    "model": "gpt-3.5-turbo",
+                    "api_key": "anything",
+                    "api_base": "http://0.0.0.0:8080",
+                },
+            },
+            {
+                "model_name": "gpt-3.5-turbo-0125-preview",
+                "litellm_params": {
+                    "model": "azure/chatgpt-v-2",
+                    "api_key": os.getenv("AZURE_API_KEY"),
+                    "api_version": os.getenv("AZURE_API_VERSION"),
+                    "api_base": os.getenv("AZURE_API_BASE"),
+                },
+            },
+        ],
+        fallbacks=[{"gpt-3.5-turbo-012": ["gpt-3.5-turbo-0125-preview"]}],
+    )
+
+    if sync_mode:
+        response = router.completion(
+            model="gpt-3.5-turbo-012",
+            messages=[{"role": "user", "content": "Hey, how's it going?"}],
+        )
+    else:
+        response = await router.acompletion(
+            model="gpt-3.5-turbo-012",
+            messages=[{"role": "user", "content": "Hey, how's it going?"}],
+        )
+
+    assert response.model == "gpt-35-turbo"
+
+
+@pytest.mark.parametrize("sync_mode", [True, False])
+@pytest.mark.parametrize("litellm_module_fallbacks", [True, False])
+@pytest.mark.asyncio
+async def test_default_model_fallbacks(sync_mode, litellm_module_fallbacks):
+    """
+    Related issue - https://github.com/BerriAI/litellm/issues/3623
+
+    If model misconfigured, setup a default model for generic fallback
+    """
+    if litellm_module_fallbacks:
+        litellm.default_fallbacks = ["my-good-model"]
+    router = Router(
+        model_list=[
+            {
+                "model_name": "bad-model",
+                "litellm_params": {
+                    "model": "openai/my-bad-model",
+                    "api_key": "my-bad-api-key",
+                },
+            },
+            {
+                "model_name": "my-good-model",
+                "litellm_params": {
+                    "model": "gpt-4o",
+                    "api_key": os.getenv("OPENAI_API_KEY"),
+                },
+            },
+        ],
+        default_fallbacks=(
+            ["my-good-model"] if litellm_module_fallbacks == False else None
+        ),
+    )
+
+    if sync_mode:
+        response = router.completion(
+            model="bad-model",
+            messages=[{"role": "user", "content": "Hey, how's it going?"}],
+            mock_testing_fallbacks=True,
+            mock_response="Hey! nice day",
+        )
+    else:
+        response = await router.acompletion(
+            model="bad-model",
+            messages=[{"role": "user", "content": "Hey, how's it going?"}],
+            mock_testing_fallbacks=True,
+            mock_response="Hey! nice day",
+        )
+
+    assert isinstance(response, litellm.ModelResponse)
+    assert response.model is not None and response.model == "gpt-4o"

litellm/tests/test_streaming.py

@@ -456,7 +456,8 @@ def test_completion_claude_stream():
         print(f"completion_response: {complete_response}")
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
-    
+
+
 # test_completion_claude_stream()
 def test_completion_claude_2_stream():
     litellm.set_verbose = True
@@ -1416,6 +1417,8 @@ def test_completion_watsonx_stream():
             raise Exception("finish reason not set for last chunk")
         if complete_response.strip() == "":
             raise Exception("Empty response received")
+    except litellm.RateLimitError as e:
+        pass
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
 

litellm/types/completion.py

@@ -1,10 +1,15 @@
 from typing import List, Optional, Union, Iterable
 
-from pydantic import BaseModel, validator
+from pydantic import ConfigDict, BaseModel, validator, VERSION
 
 from typing_extensions import Literal, Required, TypedDict
 
 
+# Function to get Pydantic version
+def is_pydantic_v2() -> int:
+    return int(VERSION.split(".")[0])
+
+
 class ChatCompletionSystemMessageParam(TypedDict, total=False):
     content: Required[str]
     """The contents of the system message."""
@@ -191,6 +196,8 @@ class CompletionRequest(BaseModel):
     api_key: Optional[str] = None
     model_list: Optional[List[str]] = None
 
-    class Config:
-        extra = "allow"
-        protected_namespaces = ()
+    # Version-specific configuration
+    if is_pydantic_v2() >= 2:
+        model_config = ConfigDict(extra="allow", protected_namespaces=())
+    else:
+        model_config = ConfigDict(extra="allow")  # No protected_namespaces for v1

litellm/types/embedding.py

@@ -1,6 +1,6 @@
 from typing import List, Optional, Union
 
-from pydantic import BaseModel, validator
+from pydantic import ConfigDict, BaseModel, validator
 
 
 class EmbeddingRequest(BaseModel):
@@ -17,7 +17,4 @@ class EmbeddingRequest(BaseModel):
     litellm_call_id: Optional[str] = None
     litellm_logging_obj: Optional[dict] = None
     logger_fn: Optional[str] = None
-
-    class Config:
-        # allow kwargs
-        extra = "allow"
+    model_config = ConfigDict(extra="allow")

litellm/types/router.py

@@ -1,6 +1,6 @@
-from typing import List, Optional, Union, Dict, Tuple, Literal
+from typing import List, Optional, Union, Dict, Tuple, Literal, TypedDict
 import httpx
-from pydantic import BaseModel, validator, Field
+from pydantic import ConfigDict, BaseModel, validator, Field, __version__ as pydantic_version
 from .completion import CompletionRequest
 from .embedding import EmbeddingRequest
 import uuid, enum
@@ -12,8 +12,9 @@ class ModelConfig(BaseModel):
     tpm: int
     rpm: int
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
 
 class RouterConfig(BaseModel):
@@ -44,8 +45,9 @@ class RouterConfig(BaseModel):
         "latency-based-routing",
     ] = "simple-shuffle"
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
 
 class UpdateRouterConfig(BaseModel):
@@ -65,8 +67,9 @@ class UpdateRouterConfig(BaseModel):
     fallbacks: Optional[List[dict]] = None
     context_window_fallbacks: Optional[List[dict]] = None
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
 
 
 class ModelInfo(BaseModel):
@@ -84,8 +87,9 @@ class ModelInfo(BaseModel):
             id = str(id)
         super().__init__(id=id, **params)
 
-    class Config:
-        extra = "allow"
+    model_config = ConfigDict(
+        extra = "allow",
+    )
 
     def __contains__(self, key):
         # Define custom behavior for the 'in' operator
@@ -180,9 +184,18 @@ class GenericLiteLLMParams(BaseModel):
             max_retries = int(max_retries)  # cast to int
         super().__init__(max_retries=max_retries, **args, **params)
 
-    class Config:
-        extra = "allow"
-        arbitrary_types_allowed = True
+    model_config = ConfigDict(
+        extra = "allow",
+        arbitrary_types_allowed = True,
+    )
+    if pydantic_version.startswith("1"):
+        # pydantic v2 warns about using a Config class.
+        # But without this, pydantic v1 will raise an error:
+        #     RuntimeError: no validator found for <class 'openai.Timeout'>,
+        #     see `arbitrary_types_allowed` in Config
+        # Putting arbitrary_types_allowed = True in the ConfigDict doesn't work in pydantic v1.
+        class Config:
+            arbitrary_types_allowed = True
 
     def __contains__(self, key):
         # Define custom behavior for the 'in' operator
@@ -241,9 +254,18 @@ class LiteLLM_Params(GenericLiteLLMParams):
             max_retries = int(max_retries)  # cast to int
         super().__init__(max_retries=max_retries, **args, **params)
 
-    class Config:
-        extra = "allow"
-        arbitrary_types_allowed = True
+    model_config = ConfigDict(
+        extra = "allow",
+        arbitrary_types_allowed = True,
+    )
+    if pydantic_version.startswith("1"):
+        # pydantic v2 warns about using a Config class.
+        # But without this, pydantic v1 will raise an error:
+        #     RuntimeError: no validator found for <class 'openai.Timeout'>,
+        #     see `arbitrary_types_allowed` in Config
+        # Putting arbitrary_types_allowed = True in the ConfigDict doesn't work in pydantic v1.
+        class Config:
+            arbitrary_types_allowed = True
 
     def __contains__(self, key):
         # Define custom behavior for the 'in' operator
@@ -273,8 +295,50 @@ class updateDeployment(BaseModel):
     litellm_params: Optional[updateLiteLLMParams] = None
     model_info: Optional[ModelInfo] = None
 
-    class Config:
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        protected_namespaces = (),
+    )
+
+
+class LiteLLMParamsTypedDict(TypedDict, total=False):
+    """
+    [TODO]
+    - allow additional params (not in list)
+    - set value to none if not set -> don't raise error if value not set
+    """
+
+    model: str
+    custom_llm_provider: Optional[str]
+    tpm: Optional[int]
+    rpm: Optional[int]
+    api_key: Optional[str]
+    api_base: Optional[str]
+    api_version: Optional[str]
+    timeout: Optional[Union[float, str, httpx.Timeout]]
+    stream_timeout: Optional[Union[float, str]]
+    max_retries: Optional[int]
+    organization: Optional[str]  # for openai orgs
+    ## UNIFIED PROJECT/REGION ##
+    region_name: Optional[str]
+    ## VERTEX AI ##
+    vertex_project: Optional[str]
+    vertex_location: Optional[str]
+    ## AWS BEDROCK / SAGEMAKER ##
+    aws_access_key_id: Optional[str]
+    aws_secret_access_key: Optional[str]
+    aws_region_name: Optional[str]
+    ## IBM WATSONX ##
+    watsonx_region_name: Optional[str]
+    ## CUSTOM PRICING ##
+    input_cost_per_token: Optional[float]
+    output_cost_per_token: Optional[float]
+    input_cost_per_second: Optional[float]
+    output_cost_per_second: Optional[float]
+
+
+class DeploymentTypedDict(TypedDict):
+    model_name: str
+    litellm_params: LiteLLMParamsTypedDict
 
 
 class Deployment(BaseModel):
@@ -307,9 +371,10 @@ class Deployment(BaseModel):
             # if using pydantic v1
             return self.dict(**kwargs)
 
-    class Config:
-        extra = "allow"
-        protected_namespaces = ()
+    model_config = ConfigDict(
+        extra = "allow",
+        protected_namespaces = (),
+    )
 
     def __contains__(self, key):
         # Define custom behavior for the 'in' operator

litellm/utils.py

@@ -6,20 +6,20 @@
 # +-----------------------------------------------+
 #
 #  Thank you users! We ❤️ you! - Krrish & Ishaan
-
 import sys, re, binascii, struct
 import litellm
 import dotenv, json, traceback, threading, base64, ast
 import subprocess, os
 from os.path import abspath, join, dirname
 import litellm, openai
+
 import itertools
 import random, uuid, requests  # type: ignore
 from functools import wraps
 import datetime, time
 import tiktoken
 import uuid
-from pydantic import BaseModel
+from pydantic import ConfigDict, BaseModel
 import aiohttp
 import textwrap
 import logging
@@ -39,21 +39,16 @@ from litellm.caching import DualCache
 oidc_cache = DualCache()
 
 try:
-    # this works in python 3.8
+    # New and recommended way to access resources
+    from importlib import resources
+
+    filename = str(resources.files(litellm).joinpath("llms/tokenizers"))
+except (ImportError, AttributeError):
+    # Old way to access resources, which setuptools deprecated some time ago
     import pkg_resources  # type: ignore
 
     filename = pkg_resources.resource_filename(__name__, "llms/tokenizers")
-# try:
-#     filename = str(
-#         resources.files().joinpath("llms/tokenizers")  # type: ignore
-#     )  # for python 3.8 and 3.12
-except:
-    # this works in python 3.9+
-    from importlib import resources
 
-    filename = str(
-        resources.files(litellm).joinpath("llms/tokenizers")  # for python 3.10
-    )  # for python 3.10+
 os.environ["TIKTOKEN_CACHE_DIR"] = (
     filename  # use local copy of tiktoken b/c of - https://github.com/BerriAI/litellm/issues/1071
 )
@@ -71,6 +66,7 @@ from .integrations.supabase import Supabase
 from .integrations.lunary import LunaryLogger
 from .integrations.prompt_layer import PromptLayerLogger
 from .integrations.langsmith import LangsmithLogger
+from .integrations.logfire_logger import LogfireLogger, LogfireLevel
 from .integrations.weights_biases import WeightsBiasesLogger
 from .integrations.custom_logger import CustomLogger
 from .integrations.langfuse import LangFuseLogger
@@ -142,6 +138,7 @@ heliconeLogger = None
 athinaLogger = None
 promptLayerLogger = None
 langsmithLogger = None
+logfireLogger = None
 weightsBiasesLogger = None
 customLogger = None
 langFuseLogger = None
@@ -330,10 +327,7 @@ class HiddenParams(OpenAIObject):
     original_response: Optional[str] = None
     model_id: Optional[str] = None  # used in Router for individual deployments
     api_base: Optional[str] = None  # returns api base used for making completion call
-
-    class Config:
-        extra = "allow"
-        protected_namespaces = ()
+    model_config = ConfigDict(extra="allow", protected_namespaces=())
 
     def get(self, key, default=None):
         # Custom .get() method to access attributes with a default value if the attribute doesn't exist
@@ -1081,7 +1075,9 @@ class CallTypes(Enum):
 
 # Logging function -> log the exact model details + what's being sent | Non-BlockingP
 class Logging:
-    global supabaseClient, liteDebuggerClient, promptLayerLogger, weightsBiasesLogger, langsmithLogger, capture_exception, add_breadcrumb, lunaryLogger
+    global supabaseClient, liteDebuggerClient, promptLayerLogger, weightsBiasesLogger, langsmithLogger, logfireLogger, capture_exception, add_breadcrumb, lunaryLogger
+
+    custom_pricing: bool = False
 
     def __init__(
         self,
@@ -1165,6 +1161,15 @@ class Logging:
             **additional_params,
         }
 
+        ## check if custom pricing set ##
+        if (
+            litellm_params.get("input_cost_per_token") is not None
+            or litellm_params.get("input_cost_per_second") is not None
+            or litellm_params.get("output_cost_per_token") is not None
+            or litellm_params.get("output_cost_per_second") is not None
+        ):
+            self.custom_pricing = True
+
     def _pre_call(self, input, api_key, model=None, additional_args={}):
         """
         Common helper function across the sync + async pre-call function
@@ -1442,10 +1447,18 @@ class Logging:
                                 )
                             )
                         else:
+                            base_model: Optional[str] = None
                             # check if base_model set on azure
                             base_model = _get_base_model_from_metadata(
                                 model_call_details=self.model_call_details
                             )
+                            # litellm model name
+                            litellm_model = self.model_call_details["model"]
+                            if (
+                                litellm_model in litellm.model_cost
+                                and self.custom_pricing == True
+                            ):
+                                base_model = litellm_model
                             # base_model defaults to None if not set on model_info
                             self.model_call_details["response_cost"] = (
                                 litellm.completion_cost(
@@ -1604,7 +1617,7 @@ class Logging:
                         # this only logs streaming once, complete_streaming_response exists i.e when stream ends
                         if self.stream:
                             if "complete_streaming_response" not in kwargs:
-                                return
+                                continue
                             else:
                                 print_verbose("reaches supabase for streaming logging!")
                                 result = kwargs["complete_streaming_response"]
@@ -1638,7 +1651,7 @@ class Logging:
                         print_verbose("reaches langsmith for logging!")
                         if self.stream:
                             if "complete_streaming_response" not in kwargs:
-                                break
+                                continue
                             else:
                                 print_verbose(
                                     "reaches langsmith for streaming logging!"
@@ -1651,6 +1664,33 @@ class Logging:
                             end_time=end_time,
                             print_verbose=print_verbose,
                         )
+                    if callback == "logfire":
+                        global logfireLogger
+                        verbose_logger.debug("reaches logfire for success logging!")
+                        kwargs = {}
+                        for k, v in self.model_call_details.items():
+                            if (
+                                k != "original_response"
+                            ):  # copy.deepcopy raises errors as this could be a coroutine
+                                kwargs[k] = v
+
+                        # this only logs streaming once, complete_streaming_response exists i.e when stream ends
+                        if self.stream:
+                            if "complete_streaming_response" not in kwargs:
+                                continue
+                            else:
+                                print_verbose("reaches logfire for streaming logging!")
+                                result = kwargs["complete_streaming_response"]
+
+                        logfireLogger.log_event(
+                            kwargs=self.model_call_details,
+                            response_obj=result,
+                            start_time=start_time,
+                            end_time=end_time,
+                            print_verbose=print_verbose,
+                            level=LogfireLevel.INFO.value,
+                        )
+
                     if callback == "lunary":
                         print_verbose("reaches lunary for logging!")
                         model = self.model
@@ -1667,7 +1707,7 @@ class Logging:
                         # this only logs streaming once, complete_streaming_response exists i.e when stream ends
                         if self.stream:
                             if "complete_streaming_response" not in kwargs:
-                                break
+                                continue
                             else:
                                 result = kwargs["complete_streaming_response"]
 
@@ -1812,7 +1852,7 @@ class Logging:
                                 f"is complete_streaming_response in kwargs: {kwargs.get('complete_streaming_response', None)}"
                             )
                             if complete_streaming_response is None:
-                                break
+                                continue
                             else:
                                 print_verbose("reaches langfuse for streaming logging!")
                                 result = kwargs["complete_streaming_response"]
@@ -1841,7 +1881,7 @@ class Logging:
                                 f"is complete_streaming_response in kwargs: {kwargs.get('complete_streaming_response', None)}"
                             )
                             if complete_streaming_response is None:
-                                break
+                                continue
                             else:
                                 print_verbose(
                                     "reaches clickhouse for streaming logging!"
@@ -1870,7 +1910,7 @@ class Logging:
                                 f"is complete_streaming_response in kwargs: {kwargs.get('complete_streaming_response', None)}"
                             )
                             if complete_streaming_response is None:
-                                break
+                                continue
                             else:
                                 print_verbose(
                                     "reaches greenscale for streaming logging!"
@@ -2341,7 +2381,9 @@ class Logging:
     def failure_handler(
         self, exception, traceback_exception, start_time=None, end_time=None
     ):
-        print_verbose(f"Logging Details LiteLLM-Failure Call")
+        print_verbose(
+            f"Logging Details LiteLLM-Failure Call: {litellm.failure_callback}"
+        )
         try:
             start_time, end_time = self._failure_handler_helper_fn(
                 exception=exception,
@@ -2396,7 +2438,7 @@ class Logging:
                             call_type=self.call_type,
                             stream=self.stream,
                         )
-                    elif callback == "lunary":
+                    if callback == "lunary":
                         print_verbose("reaches lunary for logging error!")
 
                         model = self.model
@@ -2421,7 +2463,7 @@ class Logging:
                             end_time=end_time,
                             print_verbose=print_verbose,
                         )
-                    elif callback == "sentry":
+                    if callback == "sentry":
                         print_verbose("sending exception to sentry")
                         if capture_exception:
                             capture_exception(exception)
@@ -2429,7 +2471,7 @@ class Logging:
                             print_verbose(
                                 f"capture exception not initialized: {capture_exception}"
                             )
-                    elif callable(callback):  # custom logger functions
+                    if callable(callback):  # custom logger functions
                         customLogger.log_event(
                             kwargs=self.model_call_details,
                             response_obj=result,
@@ -2438,7 +2480,7 @@ class Logging:
                             print_verbose=print_verbose,
                             callback_func=callback,
                         )
-                    elif (
+                    if (
                         isinstance(callback, CustomLogger)
                         and self.model_call_details.get("litellm_params", {}).get(
                             "acompletion", False
@@ -2455,7 +2497,7 @@ class Logging:
                             response_obj=result,
                             kwargs=self.model_call_details,
                         )
-                    elif callback == "langfuse":
+                    if callback == "langfuse":
                         global langFuseLogger
                         verbose_logger.debug("reaches langfuse for logging failure")
                         kwargs = {}
@@ -2491,7 +2533,7 @@ class Logging:
                             level="ERROR",
                             kwargs=self.model_call_details,
                         )
-                    elif callback == "prometheus":
+                    if callback == "prometheus":
                         global prometheusLogger
                         verbose_logger.debug("reaches prometheus for success logging!")
                         kwargs = {}
@@ -2509,6 +2551,26 @@ class Logging:
                             user_id=kwargs.get("user", None),
                             print_verbose=print_verbose,
                         )
+
+                    if callback == "logfire":
+                        global logfireLogger
+                        verbose_logger.debug("reaches logfire for failure logging!")
+                        kwargs = {}
+                        for k, v in self.model_call_details.items():
+                            if (
+                                k != "original_response"
+                            ):  # copy.deepcopy raises errors as this could be a coroutine
+                                kwargs[k] = v
+                        kwargs["exception"] = exception
+
+                        logfireLogger.log_event(
+                            kwargs=kwargs,
+                            response_obj=result,
+                            start_time=start_time,
+                            end_time=end_time,
+                            level=LogfireLevel.ERROR.value,
+                            print_verbose=print_verbose,
+                        )
                 except Exception as e:
                     print_verbose(
                         f"LiteLLM.LoggingError: [Non-Blocking] Exception occurred while failure logging with integrations {str(e)}"
@@ -3258,6 +3320,7 @@ def client(original_function):
                     return original_function(*args, **kwargs)
             traceback_exception = traceback.format_exc()
             end_time = datetime.datetime.now()
+
             # LOG FAILURE - handle streaming failure logging in the _next_ object, remove `handle_failure` once it's deprecated
             if logging_obj:
                 logging_obj.failure_handler(
@@ -4365,7 +4428,7 @@ def completion_cost(
     size=None,
     quality=None,
     n=None,  # number of images
-):
+) -> float:
     """
     Calculate the cost of a given completion call fot GPT-3.5-turbo, llama2, any litellm supported llm.
 
@@ -4386,10 +4449,10 @@ def completion_cost(
         - If completion_response is not provided, the function calculates token counts based on the model and input text.
         - The cost is calculated based on the model, prompt tokens, and completion tokens.
         - For certain models containing "togethercomputer" in the name, prices are based on the model size.
-        - For Replicate models, the cost is calculated based on the total time used for the request.
+        - For un-mapped Replicate models, the cost is calculated based on the total time used for the request.
 
     Exceptions:
-        - If an error occurs during execution, the function returns 0.0 without blocking the user's execution path.
+        - If an error occurs during execution, the error is raised
     """
     try:
         if (
@@ -4701,6 +4764,10 @@ def get_litellm_params(
     acompletion=None,
     preset_cache_key=None,
     no_log=None,
+    input_cost_per_second=None,
+    input_cost_per_token=None,
+    output_cost_per_token=None,
+    output_cost_per_second=None,
 ):
     litellm_params = {
         "acompletion": acompletion,
@@ -4719,6 +4786,10 @@ def get_litellm_params(
         "preset_cache_key": preset_cache_key,
         "no-log": no_log,
         "stream_response": {},  # litellm_call_id: ModelResponse Dict
+        "input_cost_per_token": input_cost_per_token,
+        "input_cost_per_second": input_cost_per_second,
+        "output_cost_per_token": output_cost_per_token,
+        "output_cost_per_second": output_cost_per_second,
     }
 
     return litellm_params
@@ -5617,32 +5688,9 @@ def get_optional_params(
             model=model, custom_llm_provider=custom_llm_provider
         )
         _check_valid_arg(supported_params=supported_params)
-        if temperature is not None:
-            optional_params["temperature"] = temperature
-        if top_p is not None:
-            optional_params["top_p"] = top_p
-        if stream is not None:
-            optional_params["stream"] = stream
-        if max_tokens is not None:
-            optional_params["max_tokens"] = max_tokens
-        if tools is not None:
-            optional_params["tools"] = tools
-        if tool_choice is not None:
-            optional_params["tool_choice"] = tool_choice
-        if response_format is not None:
-            optional_params["response_format"] = response_format
-        # check safe_mode, random_seed: https://docs.mistral.ai/api/#operation/createChatCompletion
-        safe_mode = passed_params.pop("safe_mode", None)
-        random_seed = passed_params.pop("random_seed", None)
-        extra_body = {}
-        if safe_mode is not None:
-            extra_body["safe_mode"] = safe_mode
-        if random_seed is not None:
-            extra_body["random_seed"] = random_seed
-        optional_params["extra_body"] = (
-            extra_body  # openai client supports `extra_body` param
+        optional_params = litellm.MistralConfig().map_openai_params(
+            non_default_params=non_default_params, optional_params=optional_params
         )
-
     elif custom_llm_provider == "groq":
         supported_params = get_supported_openai_params(
             model=model, custom_llm_provider=custom_llm_provider
@@ -5843,7 +5891,8 @@ def get_optional_params(
         for k in passed_params.keys():
             if k not in default_params.keys():
                 extra_body[k] = passed_params[k]
-        optional_params["extra_body"] = extra_body
+        optional_params.setdefault("extra_body", {})
+        optional_params["extra_body"] = {**optional_params["extra_body"], **extra_body}
     else:
         # if user passed in non-default kwargs for specific providers/models, pass them along
         for k in passed_params.keys():
@@ -6212,15 +6261,7 @@ def get_supported_openai_params(model: str, custom_llm_provider: str):
             "max_retries",
         ]
     elif custom_llm_provider == "mistral":
-        return [
-            "temperature",
-            "top_p",
-            "stream",
-            "max_tokens",
-            "tools",
-            "tool_choice",
-            "response_format",
-        ]
+        return litellm.MistralConfig().get_supported_openai_params()
     elif custom_llm_provider == "replicate":
         return [
             "stream",
@@ -7287,7 +7328,7 @@ def validate_environment(model: Optional[str] = None) -> dict:
 
 def set_callbacks(callback_list, function_id=None):
 
-    global sentry_sdk_instance, capture_exception, add_breadcrumb, posthog, slack_app, alerts_channel, traceloopLogger, athinaLogger, heliconeLogger, aispendLogger, berrispendLogger, supabaseClient, liteDebuggerClient, lunaryLogger, promptLayerLogger, langFuseLogger, customLogger, weightsBiasesLogger, langsmithLogger, dynamoLogger, s3Logger, dataDogLogger, prometheusLogger, greenscaleLogger, openMeterLogger
+    global sentry_sdk_instance, capture_exception, add_breadcrumb, posthog, slack_app, alerts_channel, traceloopLogger, athinaLogger, heliconeLogger, aispendLogger, berrispendLogger, supabaseClient, liteDebuggerClient, lunaryLogger, promptLayerLogger, langFuseLogger, customLogger, weightsBiasesLogger, langsmithLogger, logfireLogger, dynamoLogger, s3Logger, dataDogLogger, prometheusLogger, greenscaleLogger, openMeterLogger
 
     try:
         for callback in callback_list:
@@ -7369,6 +7410,8 @@ def set_callbacks(callback_list, function_id=None):
                 weightsBiasesLogger = WeightsBiasesLogger()
             elif callback == "langsmith":
                 langsmithLogger = LangsmithLogger()
+            elif callback == "logfire":
+                logfireLogger = LogfireLogger()
             elif callback == "aispend":
                 aispendLogger = AISpendLogger()
             elif callback == "berrispend":
@@ -7712,7 +7755,7 @@ def convert_to_model_response_object(
             for idx, choice in enumerate(response_object["choices"]):
                 message = Message(
                     content=choice["message"].get("content", None),
-                    role=choice["message"]["role"],
+                    role=choice["message"]["role"] or "assistant",
                     function_call=choice["message"].get("function_call", None),
                     tool_calls=choice["message"].get("tool_calls", None),
                 )
@@ -8516,6 +8559,15 @@ def exception_type(
                     model=model,
                     request=original_exception.request,
                 )
+            elif custom_llm_provider == "watsonx":
+                if "token_quota_reached" in error_str:
+                    exception_mapping_worked = True
+                    raise RateLimitError(
+                        message=f"WatsonxException: Rate Limit Errror - {error_str}",
+                        llm_provider="watsonx",
+                        model=model,
+                        response=original_exception.response,
+                    )
             elif custom_llm_provider == "bedrock":
                 if (
                     "too many tokens" in error_str
@@ -10761,6 +10813,8 @@ class CustomStreamWrapper:
                 else:
                     completion_obj["content"] = str(chunk)
             elif self.custom_llm_provider and (self.custom_llm_provider == "vertex_ai"):
+                import proto  # type: ignore
+
                 if self.model.startswith("claude-3"):
                     response_obj = self.handle_vertexai_anthropic_chunk(chunk=chunk)
                     if response_obj is None:
@@ -10798,10 +10852,24 @@ class CustomStreamWrapper:
                                 function_call = (
                                     chunk.candidates[0].content.parts[0].function_call
                                 )
+
                                 args_dict = {}
-                                for k, v in function_call.args.items():
-                                    args_dict[k] = v
-                                args_str = json.dumps(args_dict)
+
+                                # Check if it's a RepeatedComposite instance
+                                for key, val in function_call.args.items():
+                                    if isinstance(
+                                        val,
+                                        proto.marshal.collections.repeated.RepeatedComposite,
+                                    ):
+                                        # If so, convert to list
+                                        args_dict[key] = [v for v in val]
+                                    else:
+                                        args_dict[key] = val
+
+                                try:
+                                    args_str = json.dumps(args_dict)
+                                except Exception as e:
+                                    raise e
                                 _delta_obj = litellm.utils.Delta(
                                     content=None,
                                     tool_calls=[

model_prices_and_context_window.json

@@ -9,6 +9,30 @@
         "mode": "chat",
         "supports_function_calling": true
     },
+    "gpt-4o": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.000005,
+        "output_cost_per_token": 0.000015,
+        "litellm_provider": "openai",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true
+    },
+    "gpt-4o-2024-05-13": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.000005,
+        "output_cost_per_token": 0.000015,
+        "litellm_provider": "openai",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true
+    },
     "gpt-4-turbo-preview": {
         "max_tokens": 4096,
         "max_input_tokens": 128000,
@@ -3366,4 +3390,4 @@
         "mode": "embedding"
     }
 
-}
+}

poetry.lock

@@ -667,22 +667,46 @@ test = ["pytest (>=6)"]
 
 [[package]]
 name = "fastapi"
-version = "0.109.2"
+version = "0.111.0"
 description = "FastAPI framework, high performance, easy to learn, fast to code, ready for production"
 optional = true
 python-versions = ">=3.8"
 files = [
-    {file = "fastapi-0.109.2-py3-none-any.whl", hash = "sha256:2c9bab24667293b501cad8dd388c05240c850b58ec5876ee3283c47d6e1e3a4d"},
-    {file = "fastapi-0.109.2.tar.gz", hash = "sha256:f3817eac96fe4f65a2ebb4baa000f394e55f5fccdaf7f75250804bc58f354f73"},
+    {file = "fastapi-0.111.0-py3-none-any.whl", hash = "sha256:97ecbf994be0bcbdadedf88c3150252bed7b2087075ac99735403b1b76cc8fc0"},
+    {file = "fastapi-0.111.0.tar.gz", hash = "sha256:b9db9dd147c91cb8b769f7183535773d8741dd46f9dc6676cd82eab510228cd7"},
 ]
 
 [package.dependencies]
+email_validator = ">=2.0.0"
+fastapi-cli = ">=0.0.2"
+httpx = ">=0.23.0"
+jinja2 = ">=2.11.2"
+orjson = ">=3.2.1"
 pydantic = ">=1.7.4,<1.8 || >1.8,<1.8.1 || >1.8.1,<2.0.0 || >2.0.0,<2.0.1 || >2.0.1,<2.1.0 || >2.1.0,<3.0.0"
-starlette = ">=0.36.3,<0.37.0"
+python-multipart = ">=0.0.7"
+starlette = ">=0.37.2,<0.38.0"
 typing-extensions = ">=4.8.0"
+ujson = ">=4.0.1,<4.0.2 || >4.0.2,<4.1.0 || >4.1.0,<4.2.0 || >4.2.0,<4.3.0 || >4.3.0,<5.0.0 || >5.0.0,<5.1.0 || >5.1.0"
+uvicorn = {version = ">=0.12.0", extras = ["standard"]}
 
 [package.extras]
-all = ["email-validator (>=2.0.0)", "httpx (>=0.23.0)", "itsdangerous (>=1.1.0)", "jinja2 (>=2.11.2)", "orjson (>=3.2.1)", "pydantic-extra-types (>=2.0.0)", "pydantic-settings (>=2.0.0)", "python-multipart (>=0.0.7)", "pyyaml (>=5.3.1)", "ujson (>=4.0.1,!=4.0.2,!=4.1.0,!=4.2.0,!=4.3.0,!=5.0.0,!=5.1.0)", "uvicorn[standard] (>=0.12.0)"]
+all = ["email_validator (>=2.0.0)", "httpx (>=0.23.0)", "itsdangerous (>=1.1.0)", "jinja2 (>=2.11.2)", "orjson (>=3.2.1)", "pydantic-extra-types (>=2.0.0)", "pydantic-settings (>=2.0.0)", "python-multipart (>=0.0.7)", "pyyaml (>=5.3.1)", "ujson (>=4.0.1,!=4.0.2,!=4.1.0,!=4.2.0,!=4.3.0,!=5.0.0,!=5.1.0)", "uvicorn[standard] (>=0.12.0)"]
+
+[[package]]
+name = "fastapi-cli"
+version = "0.0.3"
+description = "Run and manage FastAPI apps from the command line with FastAPI CLI. 🚀"
+optional = true
+python-versions = ">=3.8"
+files = [
+    {file = "fastapi_cli-0.0.3-py3-none-any.whl", hash = "sha256:ae233115f729945479044917d949095e829d2d84f56f55ce1ca17627872825a5"},
+    {file = "fastapi_cli-0.0.3.tar.gz", hash = "sha256:3b6e4d2c4daee940fb8db59ebbfd60a72c4b962bcf593e263e4cc69da4ea3d7f"},
+]
+
+[package.dependencies]
+fastapi = "*"
+typer = ">=0.12.3"
+uvicorn = {version = ">=0.15.0", extras = ["standard"]}
 
 [[package]]
 name = "fastapi-sso"
@@ -1087,6 +1111,54 @@ http2 = ["h2 (>=3,<5)"]
 socks = ["socksio (==1.*)"]
 trio = ["trio (>=0.22.0,<0.26.0)"]
 
+[[package]]
+name = "httptools"
+version = "0.6.1"
+description = "A collection of framework independent HTTP protocol utils."
+optional = true
+python-versions = ">=3.8.0"
+files = [
+    {file = "httptools-0.6.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:d2f6c3c4cb1948d912538217838f6e9960bc4a521d7f9b323b3da579cd14532f"},
+    {file = "httptools-0.6.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:00d5d4b68a717765b1fabfd9ca755bd12bf44105eeb806c03d1962acd9b8e563"},
+    {file = "httptools-0.6.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:639dc4f381a870c9ec860ce5c45921db50205a37cc3334e756269736ff0aac58"},
+    {file = "httptools-0.6.1-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e57997ac7fb7ee43140cc03664de5f268813a481dff6245e0075925adc6aa185"},
+    {file = "httptools-0.6.1-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:0ac5a0ae3d9f4fe004318d64b8a854edd85ab76cffbf7ef5e32920faef62f142"},
+    {file = "httptools-0.6.1-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:3f30d3ce413088a98b9db71c60a6ada2001a08945cb42dd65a9a9fe228627658"},
+    {file = "httptools-0.6.1-cp310-cp310-win_amd64.whl", hash = "sha256:1ed99a373e327f0107cb513b61820102ee4f3675656a37a50083eda05dc9541b"},
+    {file = "httptools-0.6.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:7a7ea483c1a4485c71cb5f38be9db078f8b0e8b4c4dc0210f531cdd2ddac1ef1"},
+    {file = "httptools-0.6.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:85ed077c995e942b6f1b07583e4eb0a8d324d418954fc6af913d36db7c05a5a0"},
+    {file = "httptools-0.6.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8b0bb634338334385351a1600a73e558ce619af390c2b38386206ac6a27fecfc"},
+    {file = "httptools-0.6.1-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7d9ceb2c957320def533671fc9c715a80c47025139c8d1f3797477decbc6edd2"},
+    {file = "httptools-0.6.1-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:4f0f8271c0a4db459f9dc807acd0eadd4839934a4b9b892f6f160e94da309837"},
+    {file = "httptools-0.6.1-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:6a4f5ccead6d18ec072ac0b84420e95d27c1cdf5c9f1bc8fbd8daf86bd94f43d"},
+    {file = "httptools-0.6.1-cp311-cp311-win_amd64.whl", hash = "sha256:5cceac09f164bcba55c0500a18fe3c47df29b62353198e4f37bbcc5d591172c3"},
+    {file = "httptools-0.6.1-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:75c8022dca7935cba14741a42744eee13ba05db00b27a4b940f0d646bd4d56d0"},
+    {file = "httptools-0.6.1-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:48ed8129cd9a0d62cf4d1575fcf90fb37e3ff7d5654d3a5814eb3d55f36478c2"},
+    {file = "httptools-0.6.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6f58e335a1402fb5a650e271e8c2d03cfa7cea46ae124649346d17bd30d59c90"},
+    {file = "httptools-0.6.1-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:93ad80d7176aa5788902f207a4e79885f0576134695dfb0fefc15b7a4648d503"},
+    {file = "httptools-0.6.1-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:9bb68d3a085c2174c2477eb3ffe84ae9fb4fde8792edb7bcd09a1d8467e30a84"},
+    {file = "httptools-0.6.1-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:b512aa728bc02354e5ac086ce76c3ce635b62f5fbc32ab7082b5e582d27867bb"},
+    {file = "httptools-0.6.1-cp312-cp312-win_amd64.whl", hash = "sha256:97662ce7fb196c785344d00d638fc9ad69e18ee4bfb4000b35a52efe5adcc949"},
+    {file = "httptools-0.6.1-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:8e216a038d2d52ea13fdd9b9c9c7459fb80d78302b257828285eca1c773b99b3"},
+    {file = "httptools-0.6.1-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:3e802e0b2378ade99cd666b5bffb8b2a7cc8f3d28988685dc300469ea8dd86cb"},
+    {file = "httptools-0.6.1-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4bd3e488b447046e386a30f07af05f9b38d3d368d1f7b4d8f7e10af85393db97"},
+    {file = "httptools-0.6.1-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fe467eb086d80217b7584e61313ebadc8d187a4d95bb62031b7bab4b205c3ba3"},
+    {file = "httptools-0.6.1-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:3c3b214ce057c54675b00108ac42bacf2ab8f85c58e3f324a4e963bbc46424f4"},
+    {file = "httptools-0.6.1-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:8ae5b97f690badd2ca27cbf668494ee1b6d34cf1c464271ef7bfa9ca6b83ffaf"},
+    {file = "httptools-0.6.1-cp38-cp38-win_amd64.whl", hash = "sha256:405784577ba6540fa7d6ff49e37daf104e04f4b4ff2d1ac0469eaa6a20fde084"},
+    {file = "httptools-0.6.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:95fb92dd3649f9cb139e9c56604cc2d7c7bf0fc2e7c8d7fbd58f96e35eddd2a3"},
+    {file = "httptools-0.6.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:dcbab042cc3ef272adc11220517278519adf8f53fd3056d0e68f0a6f891ba94e"},
+    {file = "httptools-0.6.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0cf2372e98406efb42e93bfe10f2948e467edfd792b015f1b4ecd897903d3e8d"},
+    {file = "httptools-0.6.1-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:678fcbae74477a17d103b7cae78b74800d795d702083867ce160fc202104d0da"},
+    {file = "httptools-0.6.1-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:e0b281cf5a125c35f7f6722b65d8542d2e57331be573e9e88bc8b0115c4a7a81"},
+    {file = "httptools-0.6.1-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:95658c342529bba4e1d3d2b1a874db16c7cca435e8827422154c9da76ac4e13a"},
+    {file = "httptools-0.6.1-cp39-cp39-win_amd64.whl", hash = "sha256:7ebaec1bf683e4bf5e9fbb49b8cc36da482033596a415b3e4ebab5a4c0d7ec5e"},
+    {file = "httptools-0.6.1.tar.gz", hash = "sha256:c6e26c30455600b95d94b1b836085138e82f177351454ee841c148f93a9bad5a"},
+]
+
+[package.extras]
+test = ["Cython (>=0.29.24,<0.30.0)"]
+
 [[package]]
 name = "httpx"
 version = "0.27.0"
@@ -1217,6 +1289,30 @@ MarkupSafe = ">=2.0"
 [package.extras]
 i18n = ["Babel (>=2.7)"]
 
+[[package]]
+name = "markdown-it-py"
+version = "3.0.0"
+description = "Python port of markdown-it. Markdown parsing, done right!"
+optional = true
+python-versions = ">=3.8"
+files = [
+    {file = "markdown-it-py-3.0.0.tar.gz", hash = "sha256:e3f60a94fa066dc52ec76661e37c851cb232d92f9886b15cb560aaada2df8feb"},
+    {file = "markdown_it_py-3.0.0-py3-none-any.whl", hash = "sha256:355216845c60bd96232cd8d8c40e8f9765cc86f46880e43a8fd22dc1a1a8cab1"},
+]
+
+[package.dependencies]
+mdurl = ">=0.1,<1.0"
+
+[package.extras]
+benchmarking = ["psutil", "pytest", "pytest-benchmark"]
+code-style = ["pre-commit (>=3.0,<4.0)"]
+compare = ["commonmark (>=0.9,<1.0)", "markdown (>=3.4,<4.0)", "mistletoe (>=1.0,<2.0)", "mistune (>=2.0,<3.0)", "panflute (>=2.3,<3.0)"]
+linkify = ["linkify-it-py (>=1,<3)"]
+plugins = ["mdit-py-plugins"]
+profiling = ["gprof2dot"]
+rtd = ["jupyter_sphinx", "mdit-py-plugins", "myst-parser", "pyyaml", "sphinx", "sphinx-copybutton", "sphinx-design", "sphinx_book_theme"]
+testing = ["coverage", "pytest", "pytest-cov", "pytest-regressions"]
+
 [[package]]
 name = "markupsafe"
 version = "2.1.5"
@@ -1297,6 +1393,17 @@ files = [
     {file = "mccabe-0.7.0.tar.gz", hash = "sha256:348e0240c33b60bbdf4e523192ef919f28cb2c3d7d5c7794f74009290f236325"},
 ]
 
+[[package]]
+name = "mdurl"
+version = "0.1.2"
+description = "Markdown URL utilities"
+optional = true
+python-versions = ">=3.7"
+files = [
+    {file = "mdurl-0.1.2-py3-none-any.whl", hash = "sha256:84008a41e51615a49fc9966191ff91509e3c40b939176e643fd50a5c2196b8f8"},
+    {file = "mdurl-0.1.2.tar.gz", hash = "sha256:bb413d29f5eea38f31dd4754dd7377d4465116fb207585f97bf925588687c1ba"},
+]
+
 [[package]]
 name = "msal"
 version = "1.28.0"
@@ -1856,6 +1963,20 @@ files = [
     {file = "pyflakes-3.1.0.tar.gz", hash = "sha256:a0aae034c444db0071aa077972ba4768d40c830d9539fd45bf4cd3f8f6992efc"},
 ]
 
+[[package]]
+name = "pygments"
+version = "2.18.0"
+description = "Pygments is a syntax highlighting package written in Python."
+optional = true
+python-versions = ">=3.8"
+files = [
+    {file = "pygments-2.18.0-py3-none-any.whl", hash = "sha256:b8e6aca0523f3ab76fee51799c488e38782ac06eafcf95e7ba832985c8e7b13a"},
+    {file = "pygments-2.18.0.tar.gz", hash = "sha256:786ff802f32e91311bff3889f6e9a86e81505fe99f2735bb6d60ae0c5004f199"},
+]
+
+[package.extras]
+windows-terminal = ["colorama (>=0.4.6)"]
+
 [[package]]
 name = "pyjwt"
 version = "2.8.0"
@@ -2002,7 +2123,6 @@ files = [
     {file = "PyYAML-6.0.1-cp311-cp311-win_amd64.whl", hash = "sha256:bf07ee2fef7014951eeb99f56f39c9bb4af143d8aa3c21b1677805985307da34"},
     {file = "PyYAML-6.0.1-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:855fb52b0dc35af121542a76b9a84f8d1cd886ea97c84703eaa6d88e37a2ad28"},
     {file = "PyYAML-6.0.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:40df9b996c2b73138957fe23a16a4f0ba614f4c0efce1e9406a184b6d07fa3a9"},
-    {file = "PyYAML-6.0.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a08c6f0fe150303c1c6b71ebcd7213c2858041a7e01975da3a99aed1e7a378ef"},
     {file = "PyYAML-6.0.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6c22bec3fbe2524cde73d7ada88f6566758a8f7227bfbf93a408a9d86bcc12a0"},
     {file = "PyYAML-6.0.1-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:8d4e9c88387b0f5c7d5f281e55304de64cf7f9c0021a3525bd3b1c542da3b0e4"},
     {file = "PyYAML-6.0.1-cp312-cp312-win32.whl", hash = "sha256:d483d2cdf104e7c9fa60c544d92981f12ad66a457afae824d146093b8c294c54"},
@@ -2178,6 +2298,25 @@ files = [
 [package.dependencies]
 requests = "2.31.0"
 
+[[package]]
+name = "rich"
+version = "13.7.1"
+description = "Render rich text, tables, progress bars, syntax highlighting, markdown and more to the terminal"
+optional = true
+python-versions = ">=3.7.0"
+files = [
+    {file = "rich-13.7.1-py3-none-any.whl", hash = "sha256:4edbae314f59eb482f54e9e30bf00d33350aaa94f4bfcd4e9e3110e64d0d7222"},
+    {file = "rich-13.7.1.tar.gz", hash = "sha256:9be308cb1fe2f1f57d67ce99e95af38a1e2bc71ad9813b0e247cf7ffbcc3a432"},
+]
+
+[package.dependencies]
+markdown-it-py = ">=2.2.0"
+pygments = ">=2.13.0,<3.0.0"
+typing-extensions = {version = ">=4.0.0,<5.0", markers = "python_version < \"3.9\""}
+
+[package.extras]
+jupyter = ["ipywidgets (>=7.5.1,<9)"]
+
 [[package]]
 name = "rq"
 version = "1.16.2"
@@ -2223,6 +2362,17 @@ docs = ["furo", "jaraco.packaging (>=9.3)", "jaraco.tidelift (>=1.4)", "pygments
 testing = ["build[virtualenv]", "filelock (>=3.4.0)", "importlib-metadata", "ini2toml[lite] (>=0.9)", "jaraco.develop (>=7.21)", "jaraco.envs (>=2.2)", "jaraco.path (>=3.2.0)", "mypy (==1.9)", "packaging (>=23.2)", "pip (>=19.1)", "pytest (>=6,!=8.1.1)", "pytest-checkdocs (>=2.4)", "pytest-cov", "pytest-enabler (>=2.2)", "pytest-home (>=0.5)", "pytest-mypy", "pytest-perf", "pytest-ruff (>=0.2.1)", "pytest-timeout", "pytest-xdist (>=3)", "tomli", "tomli-w (>=1.0.0)", "virtualenv (>=13.0.0)", "wheel"]
 testing-integration = ["build[virtualenv] (>=1.0.3)", "filelock (>=3.4.0)", "jaraco.envs (>=2.2)", "jaraco.path (>=3.2.0)", "packaging (>=23.2)", "pytest", "pytest-enabler", "pytest-xdist", "tomli", "virtualenv (>=13.0.0)", "wheel"]
 
+[[package]]
+name = "shellingham"
+version = "1.5.4"
+description = "Tool to Detect Surrounding Shell"
+optional = true
+python-versions = ">=3.7"
+files = [
+    {file = "shellingham-1.5.4-py2.py3-none-any.whl", hash = "sha256:7ecfff8f2fd72616f7481040475a65b2bf8af90a56c89140852d1120324e8686"},
+    {file = "shellingham-1.5.4.tar.gz", hash = "sha256:8dbca0739d487e5bd35ab3ca4b36e11c4078f3a234bfce294b0a0291363404de"},
+]
+
 [[package]]
 name = "six"
 version = "1.16.0"
@@ -2247,13 +2397,13 @@ files = [
 
 [[package]]
 name = "starlette"
-version = "0.36.3"
+version = "0.37.2"
 description = "The little ASGI library that shines."
 optional = true
 python-versions = ">=3.8"
 files = [
-    {file = "starlette-0.36.3-py3-none-any.whl", hash = "sha256:13d429aa93a61dc40bf503e8c801db1f1bca3dc706b10ef2434a36123568f044"},
-    {file = "starlette-0.36.3.tar.gz", hash = "sha256:90a671733cfb35771d8cc605e0b679d23b992f8dcfad48cc60b38cb29aeb7080"},
+    {file = "starlette-0.37.2-py3-none-any.whl", hash = "sha256:6fe59f29268538e5d0d182f2791a479a0c64638e6935d1c6989e63fb2699c6ee"},
+    {file = "starlette-0.37.2.tar.gz", hash = "sha256:9af890290133b79fc3db55474ade20f6220a364a0402e0b556e7cd5e1e093823"},
 ]
 
 [package.dependencies]
@@ -2474,6 +2624,23 @@ notebook = ["ipywidgets (>=6)"]
 slack = ["slack-sdk"]
 telegram = ["requests"]
 
+[[package]]
+name = "typer"
+version = "0.12.3"
+description = "Typer, build great CLIs. Easy to code. Based on Python type hints."
+optional = true
+python-versions = ">=3.7"
+files = [
+    {file = "typer-0.12.3-py3-none-any.whl", hash = "sha256:070d7ca53f785acbccba8e7d28b08dcd88f79f1fbda035ade0aecec71ca5c914"},
+    {file = "typer-0.12.3.tar.gz", hash = "sha256:49e73131481d804288ef62598d97a1ceef3058905aa536a1134f90891ba35482"},
+]
+
+[package.dependencies]
+click = ">=8.0.0"
+rich = ">=10.11.0"
+shellingham = ">=1.3.0"
+typing-extensions = ">=3.7.4.3"
+
 [[package]]
 name = "typing-extensions"
 version = "4.11.0"
@@ -2514,6 +2681,80 @@ tzdata = {version = "*", markers = "platform_system == \"Windows\""}
 [package.extras]
 devenv = ["check-manifest", "pytest (>=4.3)", "pytest-cov", "pytest-mock (>=3.3)", "zest.releaser"]
 
+[[package]]
+name = "ujson"
+version = "5.9.0"
+description = "Ultra fast JSON encoder and decoder for Python"
+optional = true
+python-versions = ">=3.8"
+files = [
+    {file = "ujson-5.9.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:ab71bf27b002eaf7d047c54a68e60230fbd5cd9da60de7ca0aa87d0bccead8fa"},
+    {file = "ujson-5.9.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:7a365eac66f5aa7a7fdf57e5066ada6226700884fc7dce2ba5483538bc16c8c5"},
+    {file = "ujson-5.9.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e015122b337858dba5a3dc3533af2a8fc0410ee9e2374092f6a5b88b182e9fcc"},
+    {file = "ujson-5.9.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:779a2a88c53039bebfbccca934430dabb5c62cc179e09a9c27a322023f363e0d"},
+    {file = "ujson-5.9.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:10ca3c41e80509fd9805f7c149068fa8dbee18872bbdc03d7cca928926a358d5"},
+    {file = "ujson-5.9.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:4a566e465cb2fcfdf040c2447b7dd9718799d0d90134b37a20dff1e27c0e9096"},
+    {file = "ujson-5.9.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:f833c529e922577226a05bc25b6a8b3eb6c4fb155b72dd88d33de99d53113124"},
+    {file = "ujson-5.9.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:b68a0caab33f359b4cbbc10065c88e3758c9f73a11a65a91f024b2e7a1257106"},
+    {file = "ujson-5.9.0-cp310-cp310-win32.whl", hash = "sha256:7cc7e605d2aa6ae6b7321c3ae250d2e050f06082e71ab1a4200b4ae64d25863c"},
+    {file = "ujson-5.9.0-cp310-cp310-win_amd64.whl", hash = "sha256:a6d3f10eb8ccba4316a6b5465b705ed70a06011c6f82418b59278fbc919bef6f"},
+    {file = "ujson-5.9.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:3b23bbb46334ce51ddb5dded60c662fbf7bb74a37b8f87221c5b0fec1ec6454b"},
+    {file = "ujson-5.9.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:6974b3a7c17bbf829e6c3bfdc5823c67922e44ff169851a755eab79a3dd31ec0"},
+    {file = "ujson-5.9.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b5964ea916edfe24af1f4cc68488448fbb1ec27a3ddcddc2b236da575c12c8ae"},
+    {file = "ujson-5.9.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8ba7cac47dd65ff88571eceeff48bf30ed5eb9c67b34b88cb22869b7aa19600d"},
+    {file = "ujson-5.9.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:6bbd91a151a8f3358c29355a491e915eb203f607267a25e6ab10531b3b157c5e"},
+    {file = "ujson-5.9.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:829a69d451a49c0de14a9fecb2a2d544a9b2c884c2b542adb243b683a6f15908"},
+    {file = "ujson-5.9.0-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:a807ae73c46ad5db161a7e883eec0fbe1bebc6a54890152ccc63072c4884823b"},
+    {file = "ujson-5.9.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:8fc2aa18b13d97b3c8ccecdf1a3c405f411a6e96adeee94233058c44ff92617d"},
+    {file = "ujson-5.9.0-cp311-cp311-win32.whl", hash = "sha256:70e06849dfeb2548be48fdd3ceb53300640bc8100c379d6e19d78045e9c26120"},
+    {file = "ujson-5.9.0-cp311-cp311-win_amd64.whl", hash = "sha256:7309d063cd392811acc49b5016728a5e1b46ab9907d321ebbe1c2156bc3c0b99"},
+    {file = "ujson-5.9.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:20509a8c9f775b3a511e308bbe0b72897ba6b800767a7c90c5cca59d20d7c42c"},
+    {file = "ujson-5.9.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:b28407cfe315bd1b34f1ebe65d3bd735d6b36d409b334100be8cdffae2177b2f"},
+    {file = "ujson-5.9.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9d302bd17989b6bd90d49bade66943c78f9e3670407dbc53ebcf61271cadc399"},
+    {file = "ujson-5.9.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9f21315f51e0db8ee245e33a649dd2d9dce0594522de6f278d62f15f998e050e"},
+    {file = "ujson-5.9.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:5635b78b636a54a86fdbf6f027e461aa6c6b948363bdf8d4fbb56a42b7388320"},
+    {file = "ujson-5.9.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:82b5a56609f1235d72835ee109163c7041b30920d70fe7dac9176c64df87c164"},
+    {file = "ujson-5.9.0-cp312-cp312-musllinux_1_1_i686.whl", hash = "sha256:5ca35f484622fd208f55041b042d9d94f3b2c9c5add4e9af5ee9946d2d30db01"},
+    {file = "ujson-5.9.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:829b824953ebad76d46e4ae709e940bb229e8999e40881338b3cc94c771b876c"},
+    {file = "ujson-5.9.0-cp312-cp312-win32.whl", hash = "sha256:25fa46e4ff0a2deecbcf7100af3a5d70090b461906f2299506485ff31d9ec437"},
+    {file = "ujson-5.9.0-cp312-cp312-win_amd64.whl", hash = "sha256:60718f1720a61560618eff3b56fd517d107518d3c0160ca7a5a66ac949c6cf1c"},
+    {file = "ujson-5.9.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:d581db9db9e41d8ea0b2705c90518ba623cbdc74f8d644d7eb0d107be0d85d9c"},
+    {file = "ujson-5.9.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:ff741a5b4be2d08fceaab681c9d4bc89abf3c9db600ab435e20b9b6d4dfef12e"},
+    {file = "ujson-5.9.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cdcb02cabcb1e44381221840a7af04433c1dc3297af76fde924a50c3054c708c"},
+    {file = "ujson-5.9.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e208d3bf02c6963e6ef7324dadf1d73239fb7008491fdf523208f60be6437402"},
+    {file = "ujson-5.9.0-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f4b3917296630a075e04d3d07601ce2a176479c23af838b6cf90a2d6b39b0d95"},
+    {file = "ujson-5.9.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:0c4d6adb2c7bb9eb7c71ad6f6f612e13b264942e841f8cc3314a21a289a76c4e"},
+    {file = "ujson-5.9.0-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:0b159efece9ab5c01f70b9d10bbb77241ce111a45bc8d21a44c219a2aec8ddfd"},
+    {file = "ujson-5.9.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:f0cb4a7814940ddd6619bdce6be637a4b37a8c4760de9373bac54bb7b229698b"},
+    {file = "ujson-5.9.0-cp38-cp38-win32.whl", hash = "sha256:dc80f0f5abf33bd7099f7ac94ab1206730a3c0a2d17549911ed2cb6b7aa36d2d"},
+    {file = "ujson-5.9.0-cp38-cp38-win_amd64.whl", hash = "sha256:506a45e5fcbb2d46f1a51fead991c39529fc3737c0f5d47c9b4a1d762578fc30"},
+    {file = "ujson-5.9.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:d0fd2eba664a22447102062814bd13e63c6130540222c0aa620701dd01f4be81"},
+    {file = "ujson-5.9.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:bdf7fc21a03bafe4ba208dafa84ae38e04e5d36c0e1c746726edf5392e9f9f36"},
+    {file = "ujson-5.9.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e2f909bc08ce01f122fd9c24bc6f9876aa087188dfaf3c4116fe6e4daf7e194f"},
+    {file = "ujson-5.9.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bd4ea86c2afd41429751d22a3ccd03311c067bd6aeee2d054f83f97e41e11d8f"},
+    {file = "ujson-5.9.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:63fb2e6599d96fdffdb553af0ed3f76b85fda63281063f1cb5b1141a6fcd0617"},
+    {file = "ujson-5.9.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:32bba5870c8fa2a97f4a68f6401038d3f1922e66c34280d710af00b14a3ca562"},
+    {file = "ujson-5.9.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:37ef92e42535a81bf72179d0e252c9af42a4ed966dc6be6967ebfb929a87bc60"},
+    {file = "ujson-5.9.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:f69f16b8f1c69da00e38dc5f2d08a86b0e781d0ad3e4cc6a13ea033a439c4844"},
+    {file = "ujson-5.9.0-cp39-cp39-win32.whl", hash = "sha256:3382a3ce0ccc0558b1c1668950008cece9bf463ebb17463ebf6a8bfc060dae34"},
+    {file = "ujson-5.9.0-cp39-cp39-win_amd64.whl", hash = "sha256:6adef377ed583477cf005b58c3025051b5faa6b8cc25876e594afbb772578f21"},
+    {file = "ujson-5.9.0-pp310-pypy310_pp73-macosx_10_9_x86_64.whl", hash = "sha256:ffdfebd819f492e48e4f31c97cb593b9c1a8251933d8f8972e81697f00326ff1"},
+    {file = "ujson-5.9.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c4eec2ddc046360d087cf35659c7ba0cbd101f32035e19047013162274e71fcf"},
+    {file = "ujson-5.9.0-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2fbb90aa5c23cb3d4b803c12aa220d26778c31b6e4b7a13a1f49971f6c7d088e"},
+    {file = "ujson-5.9.0-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ba0823cb70866f0d6a4ad48d998dd338dce7314598721bc1b7986d054d782dfd"},
+    {file = "ujson-5.9.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:4e35d7885ed612feb6b3dd1b7de28e89baaba4011ecdf995e88be9ac614765e9"},
+    {file = "ujson-5.9.0-pp38-pypy38_pp73-macosx_10_9_x86_64.whl", hash = "sha256:b048aa93eace8571eedbd67b3766623e7f0acbf08ee291bef7d8106210432427"},
+    {file = "ujson-5.9.0-pp38-pypy38_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:323279e68c195110ef85cbe5edce885219e3d4a48705448720ad925d88c9f851"},
+    {file = "ujson-5.9.0-pp38-pypy38_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9ac92d86ff34296f881e12aa955f7014d276895e0e4e868ba7fddebbde38e378"},
+    {file = "ujson-5.9.0-pp38-pypy38_pp73-win_amd64.whl", hash = "sha256:6eecbd09b316cea1fd929b1e25f70382917542ab11b692cb46ec9b0a26c7427f"},
+    {file = "ujson-5.9.0-pp39-pypy39_pp73-macosx_10_9_x86_64.whl", hash = "sha256:473fb8dff1d58f49912323d7cb0859df5585cfc932e4b9c053bf8cf7f2d7c5c4"},
+    {file = "ujson-5.9.0-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f91719c6abafe429c1a144cfe27883eace9fb1c09a9c5ef1bcb3ae80a3076a4e"},
+    {file = "ujson-5.9.0-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7b1c0991c4fe256f5fdb19758f7eac7f47caac29a6c57d0de16a19048eb86bad"},
+    {file = "ujson-5.9.0-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2a8ea0f55a1396708e564595aaa6696c0d8af532340f477162ff6927ecc46e21"},
+    {file = "ujson-5.9.0-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:07e0cfdde5fd91f54cd2d7ffb3482c8ff1bf558abf32a8b953a5d169575ae1cd"},
+    {file = "ujson-5.9.0.tar.gz", hash = "sha256:89cc92e73d5501b8a7f48575eeb14ad27156ad092c2e9fc7e3cf949f07e75532"},
+]
+
 [[package]]
 name = "urllib3"
 version = "2.2.1"
@@ -2544,11 +2785,230 @@ files = [
 
 [package.dependencies]
 click = ">=7.0"
+colorama = {version = ">=0.4", optional = true, markers = "sys_platform == \"win32\" and extra == \"standard\""}
 h11 = ">=0.8"
+httptools = {version = ">=0.5.0", optional = true, markers = "extra == \"standard\""}
+python-dotenv = {version = ">=0.13", optional = true, markers = "extra == \"standard\""}
+pyyaml = {version = ">=5.1", optional = true, markers = "extra == \"standard\""}
+uvloop = {version = ">=0.14.0,<0.15.0 || >0.15.0,<0.15.1 || >0.15.1", optional = true, markers = "(sys_platform != \"win32\" and sys_platform != \"cygwin\") and platform_python_implementation != \"PyPy\" and extra == \"standard\""}
+watchfiles = {version = ">=0.13", optional = true, markers = "extra == \"standard\""}
+websockets = {version = ">=10.4", optional = true, markers = "extra == \"standard\""}
 
 [package.extras]
 standard = ["colorama (>=0.4)", "httptools (>=0.5.0)", "python-dotenv (>=0.13)", "pyyaml (>=5.1)", "uvloop (>=0.14.0,!=0.15.0,!=0.15.1)", "watchfiles (>=0.13)", "websockets (>=10.4)"]
 
+[[package]]
+name = "uvloop"
+version = "0.19.0"
+description = "Fast implementation of asyncio event loop on top of libuv"
+optional = true
+python-versions = ">=3.8.0"
+files = [
+    {file = "uvloop-0.19.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:de4313d7f575474c8f5a12e163f6d89c0a878bc49219641d49e6f1444369a90e"},
+    {file = "uvloop-0.19.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:5588bd21cf1fcf06bded085f37e43ce0e00424197e7c10e77afd4bbefffef428"},
+    {file = "uvloop-0.19.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7b1fd71c3843327f3bbc3237bedcdb6504fd50368ab3e04d0410e52ec293f5b8"},
+    {file = "uvloop-0.19.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5a05128d315e2912791de6088c34136bfcdd0c7cbc1cf85fd6fd1bb321b7c849"},
+    {file = "uvloop-0.19.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:cd81bdc2b8219cb4b2556eea39d2e36bfa375a2dd021404f90a62e44efaaf957"},
+    {file = "uvloop-0.19.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:5f17766fb6da94135526273080f3455a112f82570b2ee5daa64d682387fe0dcd"},
+    {file = "uvloop-0.19.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:4ce6b0af8f2729a02a5d1575feacb2a94fc7b2e983868b009d51c9a9d2149bef"},
+    {file = "uvloop-0.19.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:31e672bb38b45abc4f26e273be83b72a0d28d074d5b370fc4dcf4c4eb15417d2"},
+    {file = "uvloop-0.19.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:570fc0ed613883d8d30ee40397b79207eedd2624891692471808a95069a007c1"},
+    {file = "uvloop-0.19.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5138821e40b0c3e6c9478643b4660bd44372ae1e16a322b8fc07478f92684e24"},
+    {file = "uvloop-0.19.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:91ab01c6cd00e39cde50173ba4ec68a1e578fee9279ba64f5221810a9e786533"},
+    {file = "uvloop-0.19.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:47bf3e9312f63684efe283f7342afb414eea4d3011542155c7e625cd799c3b12"},
+    {file = "uvloop-0.19.0-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:da8435a3bd498419ee8c13c34b89b5005130a476bda1d6ca8cfdde3de35cd650"},
+    {file = "uvloop-0.19.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:02506dc23a5d90e04d4f65c7791e65cf44bd91b37f24cfc3ef6cf2aff05dc7ec"},
+    {file = "uvloop-0.19.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2693049be9d36fef81741fddb3f441673ba12a34a704e7b4361efb75cf30befc"},
+    {file = "uvloop-0.19.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7010271303961c6f0fe37731004335401eb9075a12680738731e9c92ddd96ad6"},
+    {file = "uvloop-0.19.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:5daa304d2161d2918fa9a17d5635099a2f78ae5b5960e742b2fcfbb7aefaa593"},
+    {file = "uvloop-0.19.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:7207272c9520203fea9b93843bb775d03e1cf88a80a936ce760f60bb5add92f3"},
+    {file = "uvloop-0.19.0-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:78ab247f0b5671cc887c31d33f9b3abfb88d2614b84e4303f1a63b46c046c8bd"},
+    {file = "uvloop-0.19.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:472d61143059c84947aa8bb74eabbace30d577a03a1805b77933d6bd13ddebbd"},
+    {file = "uvloop-0.19.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:45bf4c24c19fb8a50902ae37c5de50da81de4922af65baf760f7c0c42e1088be"},
+    {file = "uvloop-0.19.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:271718e26b3e17906b28b67314c45d19106112067205119dddbd834c2b7ce797"},
+    {file = "uvloop-0.19.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:34175c9fd2a4bc3adc1380e1261f60306344e3407c20a4d684fd5f3be010fa3d"},
+    {file = "uvloop-0.19.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:e27f100e1ff17f6feeb1f33968bc185bf8ce41ca557deee9d9bbbffeb72030b7"},
+    {file = "uvloop-0.19.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:13dfdf492af0aa0a0edf66807d2b465607d11c4fa48f4a1fd41cbea5b18e8e8b"},
+    {file = "uvloop-0.19.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:6e3d4e85ac060e2342ff85e90d0c04157acb210b9ce508e784a944f852a40e67"},
+    {file = "uvloop-0.19.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8ca4956c9ab567d87d59d49fa3704cf29e37109ad348f2d5223c9bf761a332e7"},
+    {file = "uvloop-0.19.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f467a5fd23b4fc43ed86342641f3936a68ded707f4627622fa3f82a120e18256"},
+    {file = "uvloop-0.19.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:492e2c32c2af3f971473bc22f086513cedfc66a130756145a931a90c3958cb17"},
+    {file = "uvloop-0.19.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:2df95fca285a9f5bfe730e51945ffe2fa71ccbfdde3b0da5772b4ee4f2e770d5"},
+    {file = "uvloop-0.19.0.tar.gz", hash = "sha256:0246f4fd1bf2bf702e06b0d45ee91677ee5c31242f39aab4ea6fe0c51aedd0fd"},
+]
+
+[package.extras]
+docs = ["Sphinx (>=4.1.2,<4.2.0)", "sphinx-rtd-theme (>=0.5.2,<0.6.0)", "sphinxcontrib-asyncio (>=0.3.0,<0.4.0)"]
+test = ["Cython (>=0.29.36,<0.30.0)", "aiohttp (==3.9.0b0)", "aiohttp (>=3.8.1)", "flake8 (>=5.0,<6.0)", "mypy (>=0.800)", "psutil", "pyOpenSSL (>=23.0.0,<23.1.0)", "pycodestyle (>=2.9.0,<2.10.0)"]
+
+[[package]]
+name = "watchfiles"
+version = "0.21.0"
+description = "Simple, modern and high performance file watching and code reload in python."
+optional = true
+python-versions = ">=3.8"
+files = [
+    {file = "watchfiles-0.21.0-cp310-cp310-macosx_10_7_x86_64.whl", hash = "sha256:27b4035013f1ea49c6c0b42d983133b136637a527e48c132d368eb19bf1ac6aa"},
+    {file = "watchfiles-0.21.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:c81818595eff6e92535ff32825f31c116f867f64ff8cdf6562cd1d6b2e1e8f3e"},
+    {file = "watchfiles-0.21.0-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:6c107ea3cf2bd07199d66f156e3ea756d1b84dfd43b542b2d870b77868c98c03"},
+    {file = "watchfiles-0.21.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0d9ac347653ebd95839a7c607608703b20bc07e577e870d824fa4801bc1cb124"},
+    {file = "watchfiles-0.21.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:5eb86c6acb498208e7663ca22dbe68ca2cf42ab5bf1c776670a50919a56e64ab"},
+    {file = "watchfiles-0.21.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f564bf68404144ea6b87a78a3f910cc8de216c6b12a4cf0b27718bf4ec38d303"},
+    {file = "watchfiles-0.21.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3d0f32ebfaa9c6011f8454994f86108c2eb9c79b8b7de00b36d558cadcedaa3d"},
+    {file = "watchfiles-0.21.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b6d45d9b699ecbac6c7bd8e0a2609767491540403610962968d258fd6405c17c"},
+    {file = "watchfiles-0.21.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:aff06b2cac3ef4616e26ba17a9c250c1fe9dd8a5d907d0193f84c499b1b6e6a9"},
+    {file = "watchfiles-0.21.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:d9792dff410f266051025ecfaa927078b94cc7478954b06796a9756ccc7e14a9"},
+    {file = "watchfiles-0.21.0-cp310-none-win32.whl", hash = "sha256:214cee7f9e09150d4fb42e24919a1e74d8c9b8a9306ed1474ecaddcd5479c293"},
+    {file = "watchfiles-0.21.0-cp310-none-win_amd64.whl", hash = "sha256:1ad7247d79f9f55bb25ab1778fd47f32d70cf36053941f07de0b7c4e96b5d235"},
+    {file = "watchfiles-0.21.0-cp311-cp311-macosx_10_7_x86_64.whl", hash = "sha256:668c265d90de8ae914f860d3eeb164534ba2e836811f91fecc7050416ee70aa7"},
+    {file = "watchfiles-0.21.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:3a23092a992e61c3a6a70f350a56db7197242f3490da9c87b500f389b2d01eef"},
+    {file = "watchfiles-0.21.0-cp311-cp311-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:e7941bbcfdded9c26b0bf720cb7e6fd803d95a55d2c14b4bd1f6a2772230c586"},
+    {file = "watchfiles-0.21.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:11cd0c3100e2233e9c53106265da31d574355c288e15259c0d40a4405cbae317"},
+    {file = "watchfiles-0.21.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:d78f30cbe8b2ce770160d3c08cff01b2ae9306fe66ce899b73f0409dc1846c1b"},
+    {file = "watchfiles-0.21.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6674b00b9756b0af620aa2a3346b01f8e2a3dc729d25617e1b89cf6af4a54eb1"},
+    {file = "watchfiles-0.21.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:fd7ac678b92b29ba630d8c842d8ad6c555abda1b9ef044d6cc092dacbfc9719d"},
+    {file = "watchfiles-0.21.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9c873345680c1b87f1e09e0eaf8cf6c891b9851d8b4d3645e7efe2ec20a20cc7"},
+    {file = "watchfiles-0.21.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:49f56e6ecc2503e7dbe233fa328b2be1a7797d31548e7a193237dcdf1ad0eee0"},
+    {file = "watchfiles-0.21.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:02d91cbac553a3ad141db016e3350b03184deaafeba09b9d6439826ee594b365"},
+    {file = "watchfiles-0.21.0-cp311-none-win32.whl", hash = "sha256:ebe684d7d26239e23d102a2bad2a358dedf18e462e8808778703427d1f584400"},
+    {file = "watchfiles-0.21.0-cp311-none-win_amd64.whl", hash = "sha256:4566006aa44cb0d21b8ab53baf4b9c667a0ed23efe4aaad8c227bfba0bf15cbe"},
+    {file = "watchfiles-0.21.0-cp311-none-win_arm64.whl", hash = "sha256:c550a56bf209a3d987d5a975cdf2063b3389a5d16caf29db4bdddeae49f22078"},
+    {file = "watchfiles-0.21.0-cp312-cp312-macosx_10_7_x86_64.whl", hash = "sha256:51ddac60b96a42c15d24fbdc7a4bfcd02b5a29c047b7f8bf63d3f6f5a860949a"},
+    {file = "watchfiles-0.21.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:511f0b034120cd1989932bf1e9081aa9fb00f1f949fbd2d9cab6264916ae89b1"},
+    {file = "watchfiles-0.21.0-cp312-cp312-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:cfb92d49dbb95ec7a07511bc9efb0faff8fe24ef3805662b8d6808ba8409a71a"},
+    {file = "watchfiles-0.21.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3f92944efc564867bbf841c823c8b71bb0be75e06b8ce45c084b46411475a915"},
+    {file = "watchfiles-0.21.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:642d66b75eda909fd1112d35c53816d59789a4b38c141a96d62f50a3ef9b3360"},
+    {file = "watchfiles-0.21.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d23bcd6c8eaa6324fe109d8cac01b41fe9a54b8c498af9ce464c1aeeb99903d6"},
+    {file = "watchfiles-0.21.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:18d5b4da8cf3e41895b34e8c37d13c9ed294954907929aacd95153508d5d89d7"},
+    {file = "watchfiles-0.21.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1b8d1eae0f65441963d805f766c7e9cd092f91e0c600c820c764a4ff71a0764c"},
+    {file = "watchfiles-0.21.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:1fd9a5205139f3c6bb60d11f6072e0552f0a20b712c85f43d42342d162be1235"},
+    {file = "watchfiles-0.21.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:a1e3014a625bcf107fbf38eece0e47fa0190e52e45dc6eee5a8265ddc6dc5ea7"},
+    {file = "watchfiles-0.21.0-cp312-none-win32.whl", hash = "sha256:9d09869f2c5a6f2d9df50ce3064b3391d3ecb6dced708ad64467b9e4f2c9bef3"},
+    {file = "watchfiles-0.21.0-cp312-none-win_amd64.whl", hash = "sha256:18722b50783b5e30a18a8a5db3006bab146d2b705c92eb9a94f78c72beb94094"},
+    {file = "watchfiles-0.21.0-cp312-none-win_arm64.whl", hash = "sha256:a3b9bec9579a15fb3ca2d9878deae789df72f2b0fdaf90ad49ee389cad5edab6"},
+    {file = "watchfiles-0.21.0-cp38-cp38-macosx_10_7_x86_64.whl", hash = "sha256:4ea10a29aa5de67de02256a28d1bf53d21322295cb00bd2d57fcd19b850ebd99"},
+    {file = "watchfiles-0.21.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:40bca549fdc929b470dd1dbfcb47b3295cb46a6d2c90e50588b0a1b3bd98f429"},
+    {file = "watchfiles-0.21.0-cp38-cp38-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:9b37a7ba223b2f26122c148bb8d09a9ff312afca998c48c725ff5a0a632145f7"},
+    {file = "watchfiles-0.21.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ec8c8900dc5c83650a63dd48c4d1d245343f904c4b64b48798c67a3767d7e165"},
+    {file = "watchfiles-0.21.0-cp38-cp38-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:8ad3fe0a3567c2f0f629d800409cd528cb6251da12e81a1f765e5c5345fd0137"},
+    {file = "watchfiles-0.21.0-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:9d353c4cfda586db2a176ce42c88f2fc31ec25e50212650c89fdd0f560ee507b"},
+    {file = "watchfiles-0.21.0-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:83a696da8922314ff2aec02987eefb03784f473281d740bf9170181829133765"},
+    {file = "watchfiles-0.21.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5a03651352fc20975ee2a707cd2d74a386cd303cc688f407296064ad1e6d1562"},
+    {file = "watchfiles-0.21.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:3ad692bc7792be8c32918c699638b660c0de078a6cbe464c46e1340dadb94c19"},
+    {file = "watchfiles-0.21.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:06247538e8253975bdb328e7683f8515ff5ff041f43be6c40bff62d989b7d0b0"},
+    {file = "watchfiles-0.21.0-cp38-none-win32.whl", hash = "sha256:9a0aa47f94ea9a0b39dd30850b0adf2e1cd32a8b4f9c7aa443d852aacf9ca214"},
+    {file = "watchfiles-0.21.0-cp38-none-win_amd64.whl", hash = "sha256:8d5f400326840934e3507701f9f7269247f7c026d1b6cfd49477d2be0933cfca"},
+    {file = "watchfiles-0.21.0-cp39-cp39-macosx_10_7_x86_64.whl", hash = "sha256:7f762a1a85a12cc3484f77eee7be87b10f8c50b0b787bb02f4e357403cad0c0e"},
+    {file = "watchfiles-0.21.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:6e9be3ef84e2bb9710f3f777accce25556f4a71e15d2b73223788d528fcc2052"},
+    {file = "watchfiles-0.21.0-cp39-cp39-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:4c48a10d17571d1275701e14a601e36959ffada3add8cdbc9e5061a6e3579a5d"},
+    {file = "watchfiles-0.21.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6c889025f59884423428c261f212e04d438de865beda0b1e1babab85ef4c0f01"},
+    {file = "watchfiles-0.21.0-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:66fac0c238ab9a2e72d026b5fb91cb902c146202bbd29a9a1a44e8db7b710b6f"},
+    {file = "watchfiles-0.21.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b4a21f71885aa2744719459951819e7bf5a906a6448a6b2bbce8e9cc9f2c8128"},
+    {file = "watchfiles-0.21.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1c9198c989f47898b2c22201756f73249de3748e0fc9de44adaf54a8b259cc0c"},
+    {file = "watchfiles-0.21.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d8f57c4461cd24fda22493109c45b3980863c58a25b8bec885ca8bea6b8d4b28"},
+    {file = "watchfiles-0.21.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:853853cbf7bf9408b404754b92512ebe3e3a83587503d766d23e6bf83d092ee6"},
+    {file = "watchfiles-0.21.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:d5b1dc0e708fad9f92c296ab2f948af403bf201db8fb2eb4c8179db143732e49"},
+    {file = "watchfiles-0.21.0-cp39-none-win32.whl", hash = "sha256:59137c0c6826bd56c710d1d2bda81553b5e6b7c84d5a676747d80caf0409ad94"},
+    {file = "watchfiles-0.21.0-cp39-none-win_amd64.whl", hash = "sha256:6cb8fdc044909e2078c248986f2fc76f911f72b51ea4a4fbbf472e01d14faa58"},
+    {file = "watchfiles-0.21.0-pp310-pypy310_pp73-macosx_10_7_x86_64.whl", hash = "sha256:ab03a90b305d2588e8352168e8c5a1520b721d2d367f31e9332c4235b30b8994"},
+    {file = "watchfiles-0.21.0-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:927c589500f9f41e370b0125c12ac9e7d3a2fd166b89e9ee2828b3dda20bfe6f"},
+    {file = "watchfiles-0.21.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1bd467213195e76f838caf2c28cd65e58302d0254e636e7c0fca81efa4a2e62c"},
+    {file = "watchfiles-0.21.0-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:02b73130687bc3f6bb79d8a170959042eb56eb3a42df3671c79b428cd73f17cc"},
+    {file = "watchfiles-0.21.0-pp38-pypy38_pp73-macosx_10_7_x86_64.whl", hash = "sha256:08dca260e85ffae975448e344834d765983237ad6dc308231aa16e7933db763e"},
+    {file = "watchfiles-0.21.0-pp38-pypy38_pp73-macosx_11_0_arm64.whl", hash = "sha256:3ccceb50c611c433145502735e0370877cced72a6c70fd2410238bcbc7fe51d8"},
+    {file = "watchfiles-0.21.0-pp38-pypy38_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:57d430f5fb63fea141ab71ca9c064e80de3a20b427ca2febcbfcef70ff0ce895"},
+    {file = "watchfiles-0.21.0-pp38-pypy38_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0dd5fad9b9c0dd89904bbdea978ce89a2b692a7ee8a0ce19b940e538c88a809c"},
+    {file = "watchfiles-0.21.0-pp39-pypy39_pp73-macosx_10_7_x86_64.whl", hash = "sha256:be6dd5d52b73018b21adc1c5d28ac0c68184a64769052dfeb0c5d9998e7f56a2"},
+    {file = "watchfiles-0.21.0-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:b3cab0e06143768499384a8a5efb9c4dc53e19382952859e4802f294214f36ec"},
+    {file = "watchfiles-0.21.0-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8c6ed10c2497e5fedadf61e465b3ca12a19f96004c15dcffe4bd442ebadc2d85"},
+    {file = "watchfiles-0.21.0-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:43babacef21c519bc6631c5fce2a61eccdfc011b4bcb9047255e9620732c8097"},
+    {file = "watchfiles-0.21.0.tar.gz", hash = "sha256:c76c635fabf542bb78524905718c39f736a98e5ab25b23ec6d4abede1a85a6a3"},
+]
+
+[package.dependencies]
+anyio = ">=3.0.0"
+
+[[package]]
+name = "websockets"
+version = "12.0"
+description = "An implementation of the WebSocket Protocol (RFC 6455 & 7692)"
+optional = true
+python-versions = ">=3.8"
+files = [
+    {file = "websockets-12.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:d554236b2a2006e0ce16315c16eaa0d628dab009c33b63ea03f41c6107958374"},
+    {file = "websockets-12.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:2d225bb6886591b1746b17c0573e29804619c8f755b5598d875bb4235ea639be"},
+    {file = "websockets-12.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:eb809e816916a3b210bed3c82fb88eaf16e8afcf9c115ebb2bacede1797d2547"},
+    {file = "websockets-12.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c588f6abc13f78a67044c6b1273a99e1cf31038ad51815b3b016ce699f0d75c2"},
+    {file = "websockets-12.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:5aa9348186d79a5f232115ed3fa9020eab66d6c3437d72f9d2c8ac0c6858c558"},
+    {file = "websockets-12.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6350b14a40c95ddd53e775dbdbbbc59b124a5c8ecd6fbb09c2e52029f7a9f480"},
+    {file = "websockets-12.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:70ec754cc2a769bcd218ed8d7209055667b30860ffecb8633a834dde27d6307c"},
+    {file = "websockets-12.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:6e96f5ed1b83a8ddb07909b45bd94833b0710f738115751cdaa9da1fb0cb66e8"},
+    {file = "websockets-12.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:4d87be612cbef86f994178d5186add3d94e9f31cc3cb499a0482b866ec477603"},
+    {file = "websockets-12.0-cp310-cp310-win32.whl", hash = "sha256:befe90632d66caaf72e8b2ed4d7f02b348913813c8b0a32fae1cc5fe3730902f"},
+    {file = "websockets-12.0-cp310-cp310-win_amd64.whl", hash = "sha256:363f57ca8bc8576195d0540c648aa58ac18cf85b76ad5202b9f976918f4219cf"},
+    {file = "websockets-12.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:5d873c7de42dea355d73f170be0f23788cf3fa9f7bed718fd2830eefedce01b4"},
+    {file = "websockets-12.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:3f61726cae9f65b872502ff3c1496abc93ffbe31b278455c418492016e2afc8f"},
+    {file = "websockets-12.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:ed2fcf7a07334c77fc8a230755c2209223a7cc44fc27597729b8ef5425aa61a3"},
+    {file = "websockets-12.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8e332c210b14b57904869ca9f9bf4ca32f5427a03eeb625da9b616c85a3a506c"},
+    {file = "websockets-12.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:5693ef74233122f8ebab026817b1b37fe25c411ecfca084b29bc7d6efc548f45"},
+    {file = "websockets-12.0-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6e9e7db18b4539a29cc5ad8c8b252738a30e2b13f033c2d6e9d0549b45841c04"},
+    {file = "websockets-12.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:6e2df67b8014767d0f785baa98393725739287684b9f8d8a1001eb2839031447"},
+    {file = "websockets-12.0-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:bea88d71630c5900690fcb03161ab18f8f244805c59e2e0dc4ffadae0a7ee0ca"},
+    {file = "websockets-12.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:dff6cdf35e31d1315790149fee351f9e52978130cef6c87c4b6c9b3baf78bc53"},
+    {file = "websockets-12.0-cp311-cp311-win32.whl", hash = "sha256:3e3aa8c468af01d70332a382350ee95f6986db479ce7af14d5e81ec52aa2b402"},
+    {file = "websockets-12.0-cp311-cp311-win_amd64.whl", hash = "sha256:25eb766c8ad27da0f79420b2af4b85d29914ba0edf69f547cc4f06ca6f1d403b"},
+    {file = "websockets-12.0-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:0e6e2711d5a8e6e482cacb927a49a3d432345dfe7dea8ace7b5790df5932e4df"},
+    {file = "websockets-12.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:dbcf72a37f0b3316e993e13ecf32f10c0e1259c28ffd0a85cee26e8549595fbc"},
+    {file = "websockets-12.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:12743ab88ab2af1d17dd4acb4645677cb7063ef4db93abffbf164218a5d54c6b"},
+    {file = "websockets-12.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7b645f491f3c48d3f8a00d1fce07445fab7347fec54a3e65f0725d730d5b99cb"},
+    {file = "websockets-12.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9893d1aa45a7f8b3bc4510f6ccf8db8c3b62120917af15e3de247f0780294b92"},
+    {file = "websockets-12.0-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1f38a7b376117ef7aff996e737583172bdf535932c9ca021746573bce40165ed"},
+    {file = "websockets-12.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:f764ba54e33daf20e167915edc443b6f88956f37fb606449b4a5b10ba42235a5"},
+    {file = "websockets-12.0-cp312-cp312-musllinux_1_1_i686.whl", hash = "sha256:1e4b3f8ea6a9cfa8be8484c9221ec0257508e3a1ec43c36acdefb2a9c3b00aa2"},
+    {file = "websockets-12.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:9fdf06fd06c32205a07e47328ab49c40fc1407cdec801d698a7c41167ea45113"},
+    {file = "websockets-12.0-cp312-cp312-win32.whl", hash = "sha256:baa386875b70cbd81798fa9f71be689c1bf484f65fd6fb08d051a0ee4e79924d"},
+    {file = "websockets-12.0-cp312-cp312-win_amd64.whl", hash = "sha256:ae0a5da8f35a5be197f328d4727dbcfafa53d1824fac3d96cdd3a642fe09394f"},
+    {file = "websockets-12.0-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:5f6ffe2c6598f7f7207eef9a1228b6f5c818f9f4d53ee920aacd35cec8110438"},
+    {file = "websockets-12.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:9edf3fc590cc2ec20dc9d7a45108b5bbaf21c0d89f9fd3fd1685e223771dc0b2"},
+    {file = "websockets-12.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:8572132c7be52632201a35f5e08348137f658e5ffd21f51f94572ca6c05ea81d"},
+    {file = "websockets-12.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:604428d1b87edbf02b233e2c207d7d528460fa978f9e391bd8aaf9c8311de137"},
+    {file = "websockets-12.0-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1a9d160fd080c6285e202327aba140fc9a0d910b09e423afff4ae5cbbf1c7205"},
+    {file = "websockets-12.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:87b4aafed34653e465eb77b7c93ef058516cb5acf3eb21e42f33928616172def"},
+    {file = "websockets-12.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:b2ee7288b85959797970114deae81ab41b731f19ebcd3bd499ae9ca0e3f1d2c8"},
+    {file = "websockets-12.0-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:7fa3d25e81bfe6a89718e9791128398a50dec6d57faf23770787ff441d851967"},
+    {file = "websockets-12.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:a571f035a47212288e3b3519944f6bf4ac7bc7553243e41eac50dd48552b6df7"},
+    {file = "websockets-12.0-cp38-cp38-win32.whl", hash = "sha256:3c6cc1360c10c17463aadd29dd3af332d4a1adaa8796f6b0e9f9df1fdb0bad62"},
+    {file = "websockets-12.0-cp38-cp38-win_amd64.whl", hash = "sha256:1bf386089178ea69d720f8db6199a0504a406209a0fc23e603b27b300fdd6892"},
+    {file = "websockets-12.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:ab3d732ad50a4fbd04a4490ef08acd0517b6ae6b77eb967251f4c263011a990d"},
+    {file = "websockets-12.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:a1d9697f3337a89691e3bd8dc56dea45a6f6d975f92e7d5f773bc715c15dde28"},
+    {file = "websockets-12.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:1df2fbd2c8a98d38a66f5238484405b8d1d16f929bb7a33ed73e4801222a6f53"},
+    {file = "websockets-12.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:23509452b3bc38e3a057382c2e941d5ac2e01e251acce7adc74011d7d8de434c"},
+    {file = "websockets-12.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2e5fc14ec6ea568200ea4ef46545073da81900a2b67b3e666f04adf53ad452ec"},
+    {file = "websockets-12.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:46e71dbbd12850224243f5d2aeec90f0aaa0f2dde5aeeb8fc8df21e04d99eff9"},
+    {file = "websockets-12.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:b81f90dcc6c85a9b7f29873beb56c94c85d6f0dac2ea8b60d995bd18bf3e2aae"},
+    {file = "websockets-12.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:a02413bc474feda2849c59ed2dfb2cddb4cd3d2f03a2fedec51d6e959d9b608b"},
+    {file = "websockets-12.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:bbe6013f9f791944ed31ca08b077e26249309639313fff132bfbf3ba105673b9"},
+    {file = "websockets-12.0-cp39-cp39-win32.whl", hash = "sha256:cbe83a6bbdf207ff0541de01e11904827540aa069293696dd528a6640bd6a5f6"},
+    {file = "websockets-12.0-cp39-cp39-win_amd64.whl", hash = "sha256:fc4e7fa5414512b481a2483775a8e8be7803a35b30ca805afa4998a84f9fd9e8"},
+    {file = "websockets-12.0-pp310-pypy310_pp73-macosx_10_9_x86_64.whl", hash = "sha256:248d8e2446e13c1d4326e0a6a4e9629cb13a11195051a73acf414812700badbd"},
+    {file = "websockets-12.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f44069528d45a933997a6fef143030d8ca8042f0dfaad753e2906398290e2870"},
+    {file = "websockets-12.0-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c4e37d36f0d19f0a4413d3e18c0d03d0c268ada2061868c1e6f5ab1a6d575077"},
+    {file = "websockets-12.0-pp310-pypy310_pp73-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3d829f975fc2e527a3ef2f9c8f25e553eb7bc779c6665e8e1d52aa22800bb38b"},
+    {file = "websockets-12.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:2c71bd45a777433dd9113847af751aae36e448bc6b8c361a566cb043eda6ec30"},
+    {file = "websockets-12.0-pp38-pypy38_pp73-macosx_10_9_x86_64.whl", hash = "sha256:0bee75f400895aef54157b36ed6d3b308fcab62e5260703add87f44cee9c82a6"},
+    {file = "websockets-12.0-pp38-pypy38_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:423fc1ed29f7512fceb727e2d2aecb952c46aa34895e9ed96071821309951123"},
+    {file = "websockets-12.0-pp38-pypy38_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:27a5e9964ef509016759f2ef3f2c1e13f403725a5e6a1775555994966a66e931"},
+    {file = "websockets-12.0-pp38-pypy38_pp73-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c3181df4583c4d3994d31fb235dc681d2aaad744fbdbf94c4802485ececdecf2"},
+    {file = "websockets-12.0-pp38-pypy38_pp73-win_amd64.whl", hash = "sha256:b067cb952ce8bf40115f6c19f478dc71c5e719b7fbaa511359795dfd9d1a6468"},
+    {file = "websockets-12.0-pp39-pypy39_pp73-macosx_10_9_x86_64.whl", hash = "sha256:00700340c6c7ab788f176d118775202aadea7602c5cc6be6ae127761c16d6b0b"},
+    {file = "websockets-12.0-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e469d01137942849cff40517c97a30a93ae79917752b34029f0ec72df6b46399"},
+    {file = "websockets-12.0-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ffefa1374cd508d633646d51a8e9277763a9b78ae71324183693959cf94635a7"},
+    {file = "websockets-12.0-pp39-pypy39_pp73-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ba0cab91b3956dfa9f512147860783a1829a8d905ee218a9837c18f683239611"},
+    {file = "websockets-12.0-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:2cb388a5bfb56df4d9a406783b7f9dbefb888c09b71629351cc6b036e9259370"},
+    {file = "websockets-12.0-py3-none-any.whl", hash = "sha256:dc284bbc8d7c78a6c69e0c7325ab46ee5e40bb4d50e494d8131a07ef47500e9e"},
+    {file = "websockets-12.0.tar.gz", hash = "sha256:81df9cbcbb6c260de1e007e58c011bfebe2dafc8435107b0537f393dd38c8b1b"},
+]
+
 [[package]]
 name = "yarl"
 version = "1.9.4"
@@ -2674,4 +3134,4 @@ proxy = ["PyJWT", "apscheduler", "backoff", "cryptography", "fastapi", "fastapi-
 [metadata]
 lock-version = "2.0"
 python-versions = ">=3.8.1,<4.0, !=3.9.7"
-content-hash = "ff38be297294f084a739ef869d41d3d80f09c80e1d05d2963073d49790f33f37"
+content-hash = "51bdb74cce68f06211fd56fb57a0293f8b43d303f31d19995bd3c452c733a9f0"

proxy_server_config.yaml

@@ -84,7 +84,6 @@ model_list:
       model: text-completion-openai/gpt-3.5-turbo-instruct
 litellm_settings:
   drop_params: True
-  enable_preview_features: True
   # max_budget: 100 
   # budget_duration: 30d
   num_retries: 5

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "litellm"
-version = "1.37.5"
+version = "1.37.9"
 description = "Library to easily interface with LLM API providers"
 authors = ["BerriAI"]
 license = "MIT"
@@ -25,7 +25,7 @@ requests = "^2.31.0"
 
 uvicorn = {version = "^0.22.0", optional = true}
 gunicorn = {version = "^22.0.0", optional = true}
-fastapi = {version = "^0.109.1", optional = true}
+fastapi = {version = "^0.111.0", optional = true}
 backoff = {version = "*", optional = true}
 pyyaml = {version = "^6.0.1", optional = true}
 rq = {version = "*", optional = true}
@@ -65,7 +65,6 @@ extra_proxy = [
     "resend"
 ]
 
-
 [tool.poetry.scripts]
 litellm = 'litellm:run_server'
 
@@ -80,7 +79,7 @@ requires = ["poetry-core", "wheel"]
 build-backend = "poetry.core.masonry.api"
 
 [tool.commitizen]
-version = "1.37.5"
+version = "1.37.9"
 version_files = [
     "pyproject.toml:^version"
 ]

requirements.txt

@@ -1,7 +1,7 @@
 # LITELLM PROXY DEPENDENCIES #
 anyio==4.2.0 # openai + http req.
 openai==1.14.3 # openai req. 
-fastapi==0.100.0 # server dep
+fastapi==0.111.0 # server dep
 backoff==2.2.1 # server dep
 pyyaml==6.0.0 # server dep
 uvicorn==0.29.0 # server dep

tests/test_openai_endpoints.py

@@ -424,10 +424,7 @@ async def test_batch_chat_completions():
         response = await chat_completion(
             session=session,
             key="sk-1234",
-            model=[
-                "gpt-3.5-turbo",
-                "fake-openai-endpoint",
-            ],
+            model="gpt-3.5-turbo,fake-openai-endpoint",
         )
 
         print(f"response: {response}")

tests/test_spend_logs.py

@@ -138,6 +138,23 @@ async def get_predict_spend_logs(session):
         return await response.json()
 
 
+async def get_spend_report(session, start_date, end_date):
+    url = "http://0.0.0.0:4000/global/spend/report"
+    headers = {"Authorization": "Bearer sk-1234", "Content-Type": "application/json"}
+    async with session.get(
+        url, headers=headers, params={"start_date": start_date, "end_date": end_date}
+    ) as response:
+        status = response.status
+        response_text = await response.text()
+
+        print(response_text)
+        print()
+
+        if status != 200:
+            raise Exception(f"Request did not return a 200 status code: {status}")
+        return await response.json()
+
+
 @pytest.mark.asyncio
 async def test_get_predicted_spend_logs():
     """
@@ -205,3 +222,39 @@ async def test_spend_logs_high_traffic():
         except:
             print(n, time.time() - start, 0)
         raise Exception("it worked!")
+
+
+@pytest.mark.asyncio
+async def test_spend_report_endpoint():
+    async with aiohttp.ClientSession(
+        timeout=aiohttp.ClientTimeout(total=600)
+    ) as session:
+        import datetime
+
+        todays_date = datetime.date.today() + datetime.timedelta(days=1)
+        todays_date = todays_date.strftime("%Y-%m-%d")
+
+        print("todays_date", todays_date)
+        thirty_days_ago = (
+            datetime.date.today() - datetime.timedelta(days=30)
+        ).strftime("%Y-%m-%d")
+        spend_report = await get_spend_report(
+            session=session, start_date=thirty_days_ago, end_date=todays_date
+        )
+        print("spend report", spend_report)
+
+        for row in spend_report:
+            date = row["group_by_day"]
+            teams = row["teams"]
+            for team in teams:
+                team_name = team["team_name"]
+                total_spend = team["total_spend"]
+                metadata = team["metadata"]
+
+                assert team_name is not None
+
+                print(f"Date: {date}")
+                print(f"Team: {team_name}")
+                print(f"Total Spend: {total_spend}")
+                print("Metadata: ", metadata)
+                print()

ui/litellm-dashboard/out/index.html

@@ -1 +1 @@
-<!DOCTYPE html><html id="__next_error__"><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width, initial-scale=1"/><link rel="preload" as="script" fetchPriority="low" href="/ui/_next/static/chunks/webpack-de9c0fadf6a94b3b.js" crossorigin=""/><script src="/ui/_next/static/chunks/fd9d1056-f960ab1e6d32b002.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/69-04708d7d4a17c1ee.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/main-app-9b4fb13a7db53edf.js" async="" crossorigin=""></script><title>LiteLLM Dashboard</title><meta name="description" content="LiteLLM Proxy Admin UI"/><link rel="icon" href="/ui/favicon.ico" type="image/x-icon" sizes="16x16"/><meta name="next-size-adjust"/><script src="/ui/_next/static/chunks/polyfills-c67a75d1b6f99dc8.js" crossorigin="" noModule=""></script></head><body><script src="/ui/_next/static/chunks/webpack-de9c0fadf6a94b3b.js" crossorigin="" async=""></script><script>(self.__next_f=self.__next_f||[]).push([0]);self.__next_f.push([2,null])</script><script>self.__next_f.push([1,"1:HL[\"/ui/_next/static/media/c9a5bc6a7c948fb0-s.p.woff2\",\"font\",{\"crossOrigin\":\"\",\"type\":\"font/woff2\"}]\n2:HL[\"/ui/_next/static/css/f04e46b02318b660.css\",\"style\",{\"crossOrigin\":\"\"}]\n0:\"$L3\"\n"])</script><script>self.__next_f.push([1,"4:I[47690,[],\"\"]\n6:I[77831,[],\"\"]\n7:I[7926,[\"936\",\"static/chunks/2f6dbc85-052c4579f80d66ae.js\",\"884\",\"static/chunks/884-7576ee407a2ecbe6.js\",\"931\",\"static/chunks/app/page-e6190351ac8da62a.js\"],\"\"]\n8:I[5613,[],\"\"]\n9:I[31778,[],\"\"]\nb:I[48955,[],\"\"]\nc:[]\n"])</script><script>self.__next_f.push([1,"3:[[[\"$\",\"link\",\"0\",{\"rel\":\"stylesheet\",\"href\":\"/ui/_next/static/css/f04e46b02318b660.css\",\"precedence\":\"next\",\"crossOrigin\":\"\"}]],[\"$\",\"$L4\",null,{\"buildId\":\"84BZ5uERcn4DsO4_POsLl\",\"assetPrefix\":\"/ui\",\"initialCanonicalUrl\":\"/\",\"initialTree\":[\"\",{\"children\":[\"__PAGE__\",{}]},\"$undefined\",\"$undefined\",true],\"initialSeedData\":[\"\",{\"children\":[\"__PAGE__\",{},[\"$L5\",[\"$\",\"$L6\",null,{\"propsForComponent\":{\"params\":{}},\"Component\":\"$7\",\"isStaticGeneration\":true}],null]]},[null,[\"$\",\"html\",null,{\"lang\":\"en\",\"children\":[\"$\",\"body\",null,{\"className\":\"__className_c23dc8\",\"children\":[\"$\",\"$L8\",null,{\"parallelRouterKey\":\"children\",\"segmentPath\":[\"children\"],\"loading\":\"$undefined\",\"loadingStyles\":\"$undefined\",\"loadingScripts\":\"$undefined\",\"hasLoading\":false,\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L9\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":[[\"$\",\"title\",null,{\"children\":\"404: This page could not be found.\"}],[\"$\",\"div\",null,{\"style\":{\"fontFamily\":\"system-ui,\\\"Segoe UI\\\",Roboto,Helvetica,Arial,sans-serif,\\\"Apple Color Emoji\\\",\\\"Segoe UI Emoji\\\"\",\"height\":\"100vh\",\"textAlign\":\"center\",\"display\":\"flex\",\"flexDirection\":\"column\",\"alignItems\":\"center\",\"justifyContent\":\"center\"},\"children\":[\"$\",\"div\",null,{\"children\":[[\"$\",\"style\",null,{\"dangerouslySetInnerHTML\":{\"__html\":\"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}\"}}],[\"$\",\"h1\",null,{\"className\":\"next-error-h1\",\"style\":{\"display\":\"inline-block\",\"margin\":\"0 20px 0 0\",\"padding\":\"0 23px 0 0\",\"fontSize\":24,\"fontWeight\":500,\"verticalAlign\":\"top\",\"lineHeight\":\"49px\"},\"children\":\"404\"}],[\"$\",\"div\",null,{\"style\":{\"display\":\"inline-block\"},\"children\":[\"$\",\"h2\",null,{\"style\":{\"fontSize\":14,\"fontWeight\":400,\"lineHeight\":\"49px\",\"margin\":0},\"children\":\"This page could not be found.\"}]}]]}]}]],\"notFoundStyles\":[],\"styles\":null}]}]}],null]],\"initialHead\":[false,\"$La\"],\"globalErrorComponent\":\"$b\",\"missingSlots\":\"$Wc\"}]]\n"])</script><script>self.__next_f.push([1,"a:[[\"$\",\"meta\",\"0\",{\"name\":\"viewport\",\"content\":\"width=device-width, initial-scale=1\"}],[\"$\",\"meta\",\"1\",{\"charSet\":\"utf-8\"}],[\"$\",\"title\",\"2\",{\"children\":\"LiteLLM Dashboard\"}],[\"$\",\"meta\",\"3\",{\"name\":\"description\",\"content\":\"LiteLLM Proxy Admin UI\"}],[\"$\",\"link\",\"4\",{\"rel\":\"icon\",\"href\":\"/ui/favicon.ico\",\"type\":\"image/x-icon\",\"sizes\":\"16x16\"}],[\"$\",\"meta\",\"5\",{\"name\":\"next-size-adjust\"}]]\n5:null\n"])</script><script>self.__next_f.push([1,""])</script></body></html>
+<!DOCTYPE html><html id="__next_error__"><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width, initial-scale=1"/><link rel="preload" as="script" fetchPriority="low" href="/ui/_next/static/chunks/webpack-de9c0fadf6a94b3b.js" crossorigin=""/><script src="/ui/_next/static/chunks/fd9d1056-f960ab1e6d32b002.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/69-04708d7d4a17c1ee.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/main-app-9b4fb13a7db53edf.js" async="" crossorigin=""></script><title>LiteLLM Dashboard</title><meta name="description" content="LiteLLM Proxy Admin UI"/><link rel="icon" href="/ui/favicon.ico" type="image/x-icon" sizes="16x16"/><meta name="next-size-adjust"/><script src="/ui/_next/static/chunks/polyfills-c67a75d1b6f99dc8.js" crossorigin="" noModule=""></script></head><body><script src="/ui/_next/static/chunks/webpack-de9c0fadf6a94b3b.js" crossorigin="" async=""></script><script>(self.__next_f=self.__next_f||[]).push([0]);self.__next_f.push([2,null])</script><script>self.__next_f.push([1,"1:HL[\"/ui/_next/static/media/c9a5bc6a7c948fb0-s.p.woff2\",\"font\",{\"crossOrigin\":\"\",\"type\":\"font/woff2\"}]\n2:HL[\"/ui/_next/static/css/f04e46b02318b660.css\",\"style\",{\"crossOrigin\":\"\"}]\n0:\"$L3\"\n"])</script><script>self.__next_f.push([1,"4:I[47690,[],\"\"]\n6:I[77831,[],\"\"]\n7:I[7926,[\"936\",\"static/chunks/2f6dbc85-052c4579f80d66ae.js\",\"884\",\"static/chunks/884-7576ee407a2ecbe6.js\",\"931\",\"static/chunks/app/page-6a39771cacf75ea6.js\"],\"\"]\n8:I[5613,[],\"\"]\n9:I[31778,[],\"\"]\nb:I[48955,[],\"\"]\nc:[]\n"])</script><script>self.__next_f.push([1,"3:[[[\"$\",\"link\",\"0\",{\"rel\":\"stylesheet\",\"href\":\"/ui/_next/static/css/f04e46b02318b660.css\",\"precedence\":\"next\",\"crossOrigin\":\"\"}]],[\"$\",\"$L4\",null,{\"buildId\":\"obp5wqVSVDMiDTC414cR8\",\"assetPrefix\":\"/ui\",\"initialCanonicalUrl\":\"/\",\"initialTree\":[\"\",{\"children\":[\"__PAGE__\",{}]},\"$undefined\",\"$undefined\",true],\"initialSeedData\":[\"\",{\"children\":[\"__PAGE__\",{},[\"$L5\",[\"$\",\"$L6\",null,{\"propsForComponent\":{\"params\":{}},\"Component\":\"$7\",\"isStaticGeneration\":true}],null]]},[null,[\"$\",\"html\",null,{\"lang\":\"en\",\"children\":[\"$\",\"body\",null,{\"className\":\"__className_c23dc8\",\"children\":[\"$\",\"$L8\",null,{\"parallelRouterKey\":\"children\",\"segmentPath\":[\"children\"],\"loading\":\"$undefined\",\"loadingStyles\":\"$undefined\",\"loadingScripts\":\"$undefined\",\"hasLoading\":false,\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L9\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":[[\"$\",\"title\",null,{\"children\":\"404: This page could not be found.\"}],[\"$\",\"div\",null,{\"style\":{\"fontFamily\":\"system-ui,\\\"Segoe UI\\\",Roboto,Helvetica,Arial,sans-serif,\\\"Apple Color Emoji\\\",\\\"Segoe UI Emoji\\\"\",\"height\":\"100vh\",\"textAlign\":\"center\",\"display\":\"flex\",\"flexDirection\":\"column\",\"alignItems\":\"center\",\"justifyContent\":\"center\"},\"children\":[\"$\",\"div\",null,{\"children\":[[\"$\",\"style\",null,{\"dangerouslySetInnerHTML\":{\"__html\":\"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}\"}}],[\"$\",\"h1\",null,{\"className\":\"next-error-h1\",\"style\":{\"display\":\"inline-block\",\"margin\":\"0 20px 0 0\",\"padding\":\"0 23px 0 0\",\"fontSize\":24,\"fontWeight\":500,\"verticalAlign\":\"top\",\"lineHeight\":\"49px\"},\"children\":\"404\"}],[\"$\",\"div\",null,{\"style\":{\"display\":\"inline-block\"},\"children\":[\"$\",\"h2\",null,{\"style\":{\"fontSize\":14,\"fontWeight\":400,\"lineHeight\":\"49px\",\"margin\":0},\"children\":\"This page could not be found.\"}]}]]}]}]],\"notFoundStyles\":[],\"styles\":null}]}]}],null]],\"initialHead\":[false,\"$La\"],\"globalErrorComponent\":\"$b\",\"missingSlots\":\"$Wc\"}]]\n"])</script><script>self.__next_f.push([1,"a:[[\"$\",\"meta\",\"0\",{\"name\":\"viewport\",\"content\":\"width=device-width, initial-scale=1\"}],[\"$\",\"meta\",\"1\",{\"charSet\":\"utf-8\"}],[\"$\",\"title\",\"2\",{\"children\":\"LiteLLM Dashboard\"}],[\"$\",\"meta\",\"3\",{\"name\":\"description\",\"content\":\"LiteLLM Proxy Admin UI\"}],[\"$\",\"link\",\"4\",{\"rel\":\"icon\",\"href\":\"/ui/favicon.ico\",\"type\":\"image/x-icon\",\"sizes\":\"16x16\"}],[\"$\",\"meta\",\"5\",{\"name\":\"next-size-adjust\"}]]\n5:null\n"])</script><script>self.__next_f.push([1,""])</script></body></html>

ui/litellm-dashboard/out/index.txt

@@ -1,7 +1,7 @@
 2:I[77831,[],""]
-3:I[7926,["936","static/chunks/2f6dbc85-052c4579f80d66ae.js","884","static/chunks/884-7576ee407a2ecbe6.js","931","static/chunks/app/page-e6190351ac8da62a.js"],""]
+3:I[7926,["936","static/chunks/2f6dbc85-052c4579f80d66ae.js","884","static/chunks/884-7576ee407a2ecbe6.js","931","static/chunks/app/page-6a39771cacf75ea6.js"],""]
 4:I[5613,[],""]
 5:I[31778,[],""]
-0:["84BZ5uERcn4DsO4_POsLl",[[["",{"children":["__PAGE__",{}]},"$undefined","$undefined",true],["",{"children":["__PAGE__",{},["$L1",["$","$L2",null,{"propsForComponent":{"params":{}},"Component":"$3","isStaticGeneration":true}],null]]},[null,["$","html",null,{"lang":"en","children":["$","body",null,{"className":"__className_c23dc8","children":["$","$L4",null,{"parallelRouterKey":"children","segmentPath":["children"],"loading":"$undefined","loadingStyles":"$undefined","loadingScripts":"$undefined","hasLoading":false,"error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L5",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":[["$","title",null,{"children":"404: This page could not be found."}],["$","div",null,{"style":{"fontFamily":"system-ui,\"Segoe UI\",Roboto,Helvetica,Arial,sans-serif,\"Apple Color Emoji\",\"Segoe UI Emoji\"","height":"100vh","textAlign":"center","display":"flex","flexDirection":"column","alignItems":"center","justifyContent":"center"},"children":["$","div",null,{"children":[["$","style",null,{"dangerouslySetInnerHTML":{"__html":"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}"}}],["$","h1",null,{"className":"next-error-h1","style":{"display":"inline-block","margin":"0 20px 0 0","padding":"0 23px 0 0","fontSize":24,"fontWeight":500,"verticalAlign":"top","lineHeight":"49px"},"children":"404"}],["$","div",null,{"style":{"display":"inline-block"},"children":["$","h2",null,{"style":{"fontSize":14,"fontWeight":400,"lineHeight":"49px","margin":0},"children":"This page could not be found."}]}]]}]}]],"notFoundStyles":[],"styles":null}]}]}],null]],[[["$","link","0",{"rel":"stylesheet","href":"/ui/_next/static/css/f04e46b02318b660.css","precedence":"next","crossOrigin":""}]],"$L6"]]]]
+0:["obp5wqVSVDMiDTC414cR8",[[["",{"children":["__PAGE__",{}]},"$undefined","$undefined",true],["",{"children":["__PAGE__",{},["$L1",["$","$L2",null,{"propsForComponent":{"params":{}},"Component":"$3","isStaticGeneration":true}],null]]},[null,["$","html",null,{"lang":"en","children":["$","body",null,{"className":"__className_c23dc8","children":["$","$L4",null,{"parallelRouterKey":"children","segmentPath":["children"],"loading":"$undefined","loadingStyles":"$undefined","loadingScripts":"$undefined","hasLoading":false,"error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L5",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":[["$","title",null,{"children":"404: This page could not be found."}],["$","div",null,{"style":{"fontFamily":"system-ui,\"Segoe UI\",Roboto,Helvetica,Arial,sans-serif,\"Apple Color Emoji\",\"Segoe UI Emoji\"","height":"100vh","textAlign":"center","display":"flex","flexDirection":"column","alignItems":"center","justifyContent":"center"},"children":["$","div",null,{"children":[["$","style",null,{"dangerouslySetInnerHTML":{"__html":"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}"}}],["$","h1",null,{"className":"next-error-h1","style":{"display":"inline-block","margin":"0 20px 0 0","padding":"0 23px 0 0","fontSize":24,"fontWeight":500,"verticalAlign":"top","lineHeight":"49px"},"children":"404"}],["$","div",null,{"style":{"display":"inline-block"},"children":["$","h2",null,{"style":{"fontSize":14,"fontWeight":400,"lineHeight":"49px","margin":0},"children":"This page could not be found."}]}]]}]}]],"notFoundStyles":[],"styles":null}]}]}],null]],[[["$","link","0",{"rel":"stylesheet","href":"/ui/_next/static/css/f04e46b02318b660.css","precedence":"next","crossOrigin":""}]],"$L6"]]]]
 6:[["$","meta","0",{"name":"viewport","content":"width=device-width, initial-scale=1"}],["$","meta","1",{"charSet":"utf-8"}],["$","title","2",{"children":"LiteLLM Dashboard"}],["$","meta","3",{"name":"description","content":"LiteLLM Proxy Admin UI"}],["$","link","4",{"rel":"icon","href":"/ui/favicon.ico","type":"image/x-icon","sizes":"16x16"}],["$","meta","5",{"name":"next-size-adjust"}]]
 1:null

ui/litellm-dashboard/package-lock.json

@@ -17,7 +17,7 @@
         "fs": "^0.0.1-security",
         "jsonwebtoken": "^9.0.2",
         "jwt-decode": "^4.0.0",
-        "next": "14.1.0",
+        "next": "14.1.1",
         "openai": "^4.28.0",
         "react": "^18",
         "react-copy-to-clipboard": "^5.1.0",
@@ -412,9 +412,9 @@
       }
     },
     "node_modules/@next/env": {
-      "version": "14.1.0",
-      "resolved": "https://registry.npmjs.org/@next/env/-/env-14.1.0.tgz",
-      "integrity": "sha512-Py8zIo+02ht82brwwhTg36iogzFqGLPXlRGKQw5s+qP/kMNc4MAyDeEwBKDijk6zTIbegEgu8Qy7C1LboslQAw=="
+      "version": "14.1.1",
+      "resolved": "https://registry.npmjs.org/@next/env/-/env-14.1.1.tgz",
+      "integrity": "sha512-7CnQyD5G8shHxQIIg3c7/pSeYFeMhsNbpU/bmvH7ZnDql7mNRgg8O2JZrhrc/soFnfBnKP4/xXNiiSIPn2w8gA=="
     },
     "node_modules/@next/eslint-plugin-next": {
       "version": "14.1.0",
@@ -426,9 +426,9 @@
       }
     },
     "node_modules/@next/swc-darwin-arm64": {
-      "version": "14.1.0",
-      "resolved": "https://registry.npmjs.org/@next/swc-darwin-arm64/-/swc-darwin-arm64-14.1.0.tgz",
-      "integrity": "sha512-nUDn7TOGcIeyQni6lZHfzNoo9S0euXnu0jhsbMOmMJUBfgsnESdjN97kM7cBqQxZa8L/bM9om/S5/1dzCrW6wQ==",
+      "version": "14.1.1",
+      "resolved": "https://registry.npmjs.org/@next/swc-darwin-arm64/-/swc-darwin-arm64-14.1.1.tgz",
+      "integrity": "sha512-yDjSFKQKTIjyT7cFv+DqQfW5jsD+tVxXTckSe1KIouKk75t1qZmj/mV3wzdmFb0XHVGtyRjDMulfVG8uCKemOQ==",
       "cpu": [
         "arm64"
       ],
@@ -441,9 +441,9 @@
       }
     },
     "node_modules/@next/swc-darwin-x64": {
-      "version": "14.1.0",
-      "resolved": "https://registry.npmjs.org/@next/swc-darwin-x64/-/swc-darwin-x64-14.1.0.tgz",
-      "integrity": "sha512-1jgudN5haWxiAl3O1ljUS2GfupPmcftu2RYJqZiMJmmbBT5M1XDffjUtRUzP4W3cBHsrvkfOFdQ71hAreNQP6g==",
+      "version": "14.1.1",
+      "resolved": "https://registry.npmjs.org/@next/swc-darwin-x64/-/swc-darwin-x64-14.1.1.tgz",
+      "integrity": "sha512-KCQmBL0CmFmN8D64FHIZVD9I4ugQsDBBEJKiblXGgwn7wBCSe8N4Dx47sdzl4JAg39IkSN5NNrr8AniXLMb3aw==",
       "cpu": [
         "x64"
       ],
@@ -456,9 +456,9 @@
       }
     },
     "node_modules/@next/swc-linux-arm64-gnu": {
-      "version": "14.1.0",
-      "resolved": "https://registry.npmjs.org/@next/swc-linux-arm64-gnu/-/swc-linux-arm64-gnu-14.1.0.tgz",
-      "integrity": "sha512-RHo7Tcj+jllXUbK7xk2NyIDod3YcCPDZxj1WLIYxd709BQ7WuRYl3OWUNG+WUfqeQBds6kvZYlc42NJJTNi4tQ==",
+      "version": "14.1.1",
+      "resolved": "https://registry.npmjs.org/@next/swc-linux-arm64-gnu/-/swc-linux-arm64-gnu-14.1.1.tgz",
+      "integrity": "sha512-YDQfbWyW0JMKhJf/T4eyFr4b3tceTorQ5w2n7I0mNVTFOvu6CGEzfwT3RSAQGTi/FFMTFcuspPec/7dFHuP7Eg==",
       "cpu": [
         "arm64"
       ],
@@ -471,9 +471,9 @@
       }
     },
     "node_modules/@next/swc-linux-arm64-musl": {
-      "version": "14.1.0",
-      "resolved": "https://registry.npmjs.org/@next/swc-linux-arm64-musl/-/swc-linux-arm64-musl-14.1.0.tgz",
-      "integrity": "sha512-v6kP8sHYxjO8RwHmWMJSq7VZP2nYCkRVQ0qolh2l6xroe9QjbgV8siTbduED4u0hlk0+tjS6/Tuy4n5XCp+l6g==",
+      "version": "14.1.1",
+      "resolved": "https://registry.npmjs.org/@next/swc-linux-arm64-musl/-/swc-linux-arm64-musl-14.1.1.tgz",
+      "integrity": "sha512-fiuN/OG6sNGRN/bRFxRvV5LyzLB8gaL8cbDH5o3mEiVwfcMzyE5T//ilMmaTrnA8HLMS6hoz4cHOu6Qcp9vxgQ==",
       "cpu": [
         "arm64"
       ],
@@ -486,9 +486,9 @@
       }
     },
     "node_modules/@next/swc-linux-x64-gnu": {
-      "version": "14.1.0",
-      "resolved": "https://registry.npmjs.org/@next/swc-linux-x64-gnu/-/swc-linux-x64-gnu-14.1.0.tgz",
-      "integrity": "sha512-zJ2pnoFYB1F4vmEVlb/eSe+VH679zT1VdXlZKX+pE66grOgjmKJHKacf82g/sWE4MQ4Rk2FMBCRnX+l6/TVYzQ==",
+      "version": "14.1.1",
+      "resolved": "https://registry.npmjs.org/@next/swc-linux-x64-gnu/-/swc-linux-x64-gnu-14.1.1.tgz",
+      "integrity": "sha512-rv6AAdEXoezjbdfp3ouMuVqeLjE1Bin0AuE6qxE6V9g3Giz5/R3xpocHoAi7CufRR+lnkuUjRBn05SYJ83oKNQ==",
       "cpu": [
         "x64"
       ],
@@ -501,9 +501,9 @@
       }
     },
     "node_modules/@next/swc-linux-x64-musl": {
-      "version": "14.1.0",
-      "resolved": "https://registry.npmjs.org/@next/swc-linux-x64-musl/-/swc-linux-x64-musl-14.1.0.tgz",
-      "integrity": "sha512-rbaIYFt2X9YZBSbH/CwGAjbBG2/MrACCVu2X0+kSykHzHnYH5FjHxwXLkcoJ10cX0aWCEynpu+rP76x0914atg==",
+      "version": "14.1.1",
+      "resolved": "https://registry.npmjs.org/@next/swc-linux-x64-musl/-/swc-linux-x64-musl-14.1.1.tgz",
+      "integrity": "sha512-YAZLGsaNeChSrpz/G7MxO3TIBLaMN8QWMr3X8bt6rCvKovwU7GqQlDu99WdvF33kI8ZahvcdbFsy4jAFzFX7og==",
       "cpu": [
         "x64"
       ],
@@ -516,9 +516,9 @@
       }
     },
     "node_modules/@next/swc-win32-arm64-msvc": {
-      "version": "14.1.0",
-      "resolved": "https://registry.npmjs.org/@next/swc-win32-arm64-msvc/-/swc-win32-arm64-msvc-14.1.0.tgz",
-      "integrity": "sha512-o1N5TsYc8f/HpGt39OUQpQ9AKIGApd3QLueu7hXk//2xq5Z9OxmV6sQfNp8C7qYmiOlHYODOGqNNa0e9jvchGQ==",
+      "version": "14.1.1",
+      "resolved": "https://registry.npmjs.org/@next/swc-win32-arm64-msvc/-/swc-win32-arm64-msvc-14.1.1.tgz",
+      "integrity": "sha512-1L4mUYPBMvVDMZg1inUYyPvFSduot0g73hgfD9CODgbr4xiTYe0VOMTZzaRqYJYBA9mana0x4eaAaypmWo1r5A==",
       "cpu": [
         "arm64"
       ],
@@ -531,9 +531,9 @@
       }
     },
     "node_modules/@next/swc-win32-ia32-msvc": {
-      "version": "14.1.0",
-      "resolved": "https://registry.npmjs.org/@next/swc-win32-ia32-msvc/-/swc-win32-ia32-msvc-14.1.0.tgz",
-      "integrity": "sha512-XXIuB1DBRCFwNO6EEzCTMHT5pauwaSj4SWs7CYnME57eaReAKBXCnkUE80p/pAZcewm7hs+vGvNqDPacEXHVkw==",
+      "version": "14.1.1",
+      "resolved": "https://registry.npmjs.org/@next/swc-win32-ia32-msvc/-/swc-win32-ia32-msvc-14.1.1.tgz",
+      "integrity": "sha512-jvIE9tsuj9vpbbXlR5YxrghRfMuG0Qm/nZ/1KDHc+y6FpnZ/apsgh+G6t15vefU0zp3WSpTMIdXRUsNl/7RSuw==",
       "cpu": [
         "ia32"
       ],
@@ -546,9 +546,9 @@
       }
     },
     "node_modules/@next/swc-win32-x64-msvc": {
-      "version": "14.1.0",
-      "resolved": "https://registry.npmjs.org/@next/swc-win32-x64-msvc/-/swc-win32-x64-msvc-14.1.0.tgz",
-      "integrity": "sha512-9WEbVRRAqJ3YFVqEZIxUqkiO8l1nool1LmNxygr5HWF8AcSYsEpneUDhmjUVJEzO2A04+oPtZdombzzPPkTtgg==",
+      "version": "14.1.1",
+      "resolved": "https://registry.npmjs.org/@next/swc-win32-x64-msvc/-/swc-win32-x64-msvc-14.1.1.tgz",
+      "integrity": "sha512-S6K6EHDU5+1KrBDLko7/c1MNy/Ya73pIAmvKeFwsF4RmBFJSO7/7YeD4FnZ4iBdzE69PpQ4sOMU9ORKeNuxe8A==",
       "cpu": [
         "x64"
       ],
@@ -4907,11 +4907,11 @@
       "dev": true
     },
     "node_modules/next": {
-      "version": "14.1.0",
-      "resolved": "https://registry.npmjs.org/next/-/next-14.1.0.tgz",
-      "integrity": "sha512-wlzrsbfeSU48YQBjZhDzOwhWhGsy+uQycR8bHAOt1LY1bn3zZEcDyHQOEoN3aWzQ8LHCAJ1nqrWCc9XF2+O45Q==",
+      "version": "14.1.1",
+      "resolved": "https://registry.npmjs.org/next/-/next-14.1.1.tgz",
+      "integrity": "sha512-McrGJqlGSHeaz2yTRPkEucxQKe5Zq7uPwyeHNmJaZNY4wx9E9QdxmTp310agFRoMuIYgQrCrT3petg13fSVOww==",
       "dependencies": {
-        "@next/env": "14.1.0",
+        "@next/env": "14.1.1",
         "@swc/helpers": "0.5.2",
         "busboy": "1.6.0",
         "caniuse-lite": "^1.0.30001579",
@@ -4926,15 +4926,15 @@
         "node": ">=18.17.0"
       },
       "optionalDependencies": {
-        "@next/swc-darwin-arm64": "14.1.0",
-        "@next/swc-darwin-x64": "14.1.0",
-        "@next/swc-linux-arm64-gnu": "14.1.0",
-        "@next/swc-linux-arm64-musl": "14.1.0",
-        "@next/swc-linux-x64-gnu": "14.1.0",
-        "@next/swc-linux-x64-musl": "14.1.0",
-        "@next/swc-win32-arm64-msvc": "14.1.0",
-        "@next/swc-win32-ia32-msvc": "14.1.0",
-        "@next/swc-win32-x64-msvc": "14.1.0"
+        "@next/swc-darwin-arm64": "14.1.1",
+        "@next/swc-darwin-x64": "14.1.1",
+        "@next/swc-linux-arm64-gnu": "14.1.1",
+        "@next/swc-linux-arm64-musl": "14.1.1",
+        "@next/swc-linux-x64-gnu": "14.1.1",
+        "@next/swc-linux-x64-musl": "14.1.1",
+        "@next/swc-win32-arm64-msvc": "14.1.1",
+        "@next/swc-win32-ia32-msvc": "14.1.1",
+        "@next/swc-win32-x64-msvc": "14.1.1"
       },
       "peerDependencies": {
         "@opentelemetry/api": "^1.1.0",

ui/litellm-dashboard/package.json

@@ -18,7 +18,7 @@
     "fs": "^0.0.1-security",
     "jsonwebtoken": "^9.0.2",
     "jwt-decode": "^4.0.0",
-    "next": "14.1.0",
+    "next": "14.1.1",
     "openai": "^4.28.0",
     "react": "^18",
     "react-copy-to-clipboard": "^5.1.0",

ui/litellm-dashboard/src/components/networking.tsx

@@ -655,11 +655,20 @@ export const teamSpendLogsCall = async (accessToken: String) => {
 };
 
 
-export const tagsSpendLogsCall = async (accessToken: String) => {
+export const tagsSpendLogsCall = async (
+    accessToken: String,
+    startTime: String | undefined,
+    endTime: String | undefined
+  ) => {
   try {
-    const url = proxyBaseUrl
+    let url = proxyBaseUrl
       ? `${proxyBaseUrl}/global/spend/tags`
       : `/global/spend/tags`;
+    
+    if (startTime && endTime) {
+      url = `${url}?start_date=${startTime}&end_date=${endTime}`
+    }
+
     console.log("in tagsSpendLogsCall:", url);
     const response = await fetch(`${url}`, {
       method: "GET",

ui/litellm-dashboard/src/components/settings.tsx

@@ -109,6 +109,7 @@ const Settings: React.FC<SettingsPageProps> = ({
     "llm_requests_hanging": "LLM Requests Hanging",
     "budget_alerts": "Budget Alerts (API Keys, Users)",
     "db_exceptions": "Database Exceptions (Read/Write)",
+    "daily_reports": "Weekly/Monthly Spend Reports",
   }
 
   useEffect(() => {

ui/litellm-dashboard/src/components/teams.tsx

@@ -129,7 +129,7 @@ const Team: React.FC<TeamProps> = ({
               name="team_alias"
               rules={[{ required: true, message: "Please input a team name" }]}
             >
-              <Input />
+              <TextInput />
             </Form.Item>
             <Form.Item label="Models" name="models">
               <Select2

ui/litellm-dashboard/src/components/usage.tsx

@@ -153,6 +153,19 @@ const UsagePage: React.FC<UsagePageProps> = ({
     console.log("End user data updated successfully", newTopUserData);
     setTopUsers(newTopUserData);
   
+  }
+
+  const updateTagSpendData = async (startTime:  Date | undefined, endTime:  Date | undefined) => {
+    if (!startTime || !endTime || !accessToken) {
+      return;
+    }
+
+    let top_tags = await tagsSpendLogsCall(accessToken, startTime.toISOString(), endTime.toISOString());
+    setTopTagsData(top_tags.spend_per_tag);
+    console.log("Tag spend data updated successfully");
+
+
+
   }
 
   function formatDate(date: Date) {
@@ -218,8 +231,8 @@ const UsagePage: React.FC<UsagePageProps> = ({
             setTotalSpendPerTeam(total_spend_per_team);
 
             //get top tags
-            const top_tags = await tagsSpendLogsCall(accessToken);
-            setTopTagsData(top_tags.top_10_tags);
+            const top_tags = await tagsSpendLogsCall(accessToken, dateValue.from?.toISOString(), dateValue.to?.toISOString());
+            setTopTagsData(top_tags.spend_per_tag);
 
             // get spend per end-user
             let spend_user_call = await adminTopEndUsersCall(accessToken, null, undefined, undefined);
@@ -459,38 +472,28 @@ const UsagePage: React.FC<UsagePageProps> = ({
             <TabPanel>
             <Grid numItems={2} className="gap-2 h-[75vh] w-full mb-4">
             <Col numColSpan={2}>
+            <DateRangePicker 
+                  className="mb-4"
+                  enableSelect={true} 
+                  value={dateValue} 
+                  onValueChange={(value) => {
+                    setDateValue(value);
+                    updateTagSpendData(value.from, value.to); // Call updateModelMetrics with the new date range
+                  }}
+              />
 
               <Card>
-              <Title>Spend Per Tag - Last 30 Days</Title>
-              <Text>Get Started Tracking cost per tag <a href="https://docs.litellm.ai/docs/proxy/enterprise#tracking-spend-for-custom-tags" target="_blank">here</a></Text>
-              <Table>
-                <TableHead>
-                  <TableRow>
-                    <TableHeaderCell>Tag</TableHeaderCell>
-                    <TableHeaderCell>Spend</TableHeaderCell>
-                    <TableHeaderCell>Requests</TableHeaderCell>
-                  </TableRow>
-                </TableHead>
-                <TableBody>
-                  {topTagsData.map((tag) => (
-                    <TableRow key={tag.name}>
-                      <TableCell>{tag.name}</TableCell>
-                      <TableCell>{tag.value}</TableCell>
-                      <TableCell>{tag.log_count}</TableCell>
-                    </TableRow>
-                  ))}
-                </TableBody>
-              </Table>
-                {/* <BarChart
-                  className="h-72"
-                  data={teamSpendData}
-                  showLegend={true}
-                  index="date"
-                  categories={uniqueTeamIds}
-                  yAxisWidth={80}
-                  
-                  stack={true}
-                /> */}
+              <Title>Spend Per Tag</Title>
+              <Text>Get Started Tracking cost per tag <a className="text-blue-500" href="https://docs.litellm.ai/docs/proxy/enterprise#tracking-spend-for-custom-tags" target="_blank">here</a></Text>
+             <BarChart
+              className="h-72"
+              data={topTagsData}
+              index="name"
+              categories={["spend"]}
+              colors={["blue"]}
+             >
+
+             </BarChart>
               </Card>
               </Col>
               <Col numColSpan={2}>