Commit Graph

11426 Commits

Author SHA1 Message Date
Ishaan Jaff efaf4c87f2 Merge pull request #3601 from msabramo/msabramo/update-fastapi-to-update-starlette-to-fix-httpx-warnings
Update FastAPI to update starlette to fix warnings
2024-05-13 12:20:17 -07:00
Krrish Dholakia 5dc3f157a6 build(model_prices_and_context_window.json): fix gpt-4o max tokens 2024-05-13 12:11:15 -07:00
Krish Dholakia 6c2168b6f7 Merge pull request #3613 from ConnorDoyle/patch-1
Add gpt-4o metadata
2024-05-13 12:09:27 -07:00
Connor Doyle 5da5021918 Update model_prices_and_context_window.json 2024-05-13 11:51:54 -07:00
Connor Doyle a4380f67a9 Add gpt-4o metadata
#3612
2024-05-13 11:12:17 -07:00
Krrish Dholakia 04ae285001 fix(vertex_ai.py): support tool call list response async completion 2024-05-13 10:42:31 -07:00
Krrish Dholakia 7f6e933372 fix(router.py): give an 'info' log when fallbacks work successfully 2024-05-13 10:17:32 -07:00
Krrish Dholakia 13e1577753 fix(slack_alerting.py): don't fire spam alerts when backend api call fails 2024-05-13 10:04:43 -07:00
Marc Abramowitz e73e36f210 requirements.txt: Update fastapi to 0.111.0 2024-05-13 09:21:19 -07:00
Krrish Dholakia b063ef7a47 bump: version 1.37.5 → 1.37.6 2024-05-13 09:08:04 -07:00
Krrish Dholakia 5342b3dc05 fix(router.py): fix error message to return if pre-call-checks + allowed model region 2024-05-13 09:04:38 -07:00
Krrish Dholakia c3293474dd fix(proxy_server.py): return 'allowed-model-region' in headers 2024-05-13 08:48:16 -07:00
Ishaan Jaff 514c5737f8 Merge pull request #3587 from BerriAI/litellm_proxy_use_batch_completions_model_csv
[Feat] Use csv values for proxy batch completions (OpenAI Python compatible)
2024-05-13 07:55:12 -07:00
Marc Abramowitz 82f43f98b3 Update FastAPI to update starlette to fix warnings
The httpx package emits some warnings when we run the tests in
`test_proxy_server.py`:

```shell
$ env -i PATH=$PATH poetry run pytest litellm/tests/test_proxy_server.py
...
litellm/tests/test_proxy_server.py::test_bedrock_embedding
litellm/tests/test_proxy_server.py::test_chat_completion
litellm/tests/test_proxy_server.py::test_chat_completion_azure
litellm/tests/test_proxy_server.py::test_chat_completion_optional_params
litellm/tests/test_proxy_server.py::test_embedding
litellm/tests/test_proxy_server.py::test_engines_model_chat_completions
litellm/tests/test_proxy_server.py::test_health
litellm/tests/test_proxy_server.py::test_img_gen
litellm/tests/test_proxy_server.py::test_openai_deployments_model_chat_completions_azure
  .../site-packages/httpx/_client.py:680:
  DeprecationWarning: The 'app' shortcut is now deprecated.
  Use the explicit style 'transport=WSGITransport(app=...)' instead.
    warnings.warn(message, DeprecationWarning)
...
```

Fixing this requires updating starlette, but the version of FastAPI in use
prevents that. So I updated FastAPI to update starlette to fix the httpx
warnings.

When run in combination with the changes in GH-3600, the tests pass and only
emit 1 warning!

```shell
$ env -i PATH=$PATH poetry run pytest litellm/tests/test_proxy_server.py
====================================== test session starts ======================================
platform darwin -- Python 3.12.3, pytest-7.4.4, pluggy-1.5.0
rootdir: /Users/abramowi/Code/OpenSource/litellm
plugins: anyio-4.3.0, mock-3.14.0
collected 12 items

litellm/tests/test_proxy_server.py s..........s                                           [100%]

======================================= warnings summary ========================================
litellm/utils.py:43
  /Users/abramowi/Code/OpenSource/litellm/litellm/utils.py:43: DeprecationWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html
    import pkg_resources  # type: ignore

-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
=========================== 10 passed, 2 skipped, 1 warning in 8.81s ============================
```
2024-05-12 12:26:23 -07:00
Krrish Dholakia 61143c8b45 refactor(main.py): trigger new build v1.37.5 v1.37.5-stable 2024-05-11 22:53:09 -07:00
Krrish Dholakia 99e8f0715e test(test_end_users.py): fix end user region routing test 2024-05-11 22:42:43 -07:00
Krrish Dholakia b4684d5132 fix(proxy_server.py): linting fix 2024-05-11 22:05:01 -07:00
Krrish Dholakia 094f20121a build(model_prices_and_context_window.json): add bedrock cohere command r pricing 2024-05-11 21:38:53 -07:00
Krrish Dholakia 15a6e59431 fix(proxy/_types.py): allow jwt admin to access spend routes 2024-05-11 21:31:34 -07:00
Krrish Dholakia 7276c6eb1e docs(token_auth.md): add end user cost tracking to jwt auth docs 2024-05-11 21:28:31 -07:00
Krish Dholakia 94c9df969e Merge pull request #3586 from BerriAI/litellm_bedrock_command_r_support
feat(bedrock_httpx.py): Make Bedrock-Cohere calls Async
2024-05-11 21:24:51 -07:00
Krish Dholakia 1d651c6049 Merge branch 'main' into litellm_bedrock_command_r_support 2024-05-11 21:24:42 -07:00
Krrish Dholakia e8437e52fa test(test_rules.py): fix test 2024-05-11 21:22:37 -07:00
Krish Dholakia 7566a2fc78 Merge pull request #3589 from msabramo/msabramo/make_test_load_router_config_pass
Make `test_load_router_config` pass
2024-05-11 21:15:07 -07:00
Krrish Dholakia d142478b75 fix(langfuse.py): fix handling of dict object for langfuse prompt management 2024-05-11 20:42:55 -07:00
Ishaan Jaff d3371fc81d fix langfuse logging metadata 2024-05-11 20:39:44 -07:00
Ishaan Jaff 2b3414c667 ci/cd run again 2024-05-11 20:34:55 -07:00
Krrish Dholakia 15ba244e46 fix(utils.py): correctly exception map 'request too large' as rate limit error 2024-05-11 20:20:34 -07:00
Krrish Dholakia a456f6bf2b fix(anthropic.py): fix tool calling + streaming issue 2024-05-11 20:15:36 -07:00
Krrish Dholakia 83beb41096 fix(anthropic_text.py): fix linting error 2024-05-11 20:01:50 -07:00
Ishaan Jaff beac60ed12 test - router retry policy 2024-05-11 19:58:17 -07:00
Krrish Dholakia 65d0be85fc fix(bedrock_httpx.py): compatibility fix 2024-05-11 19:55:38 -07:00
Krrish Dholakia f6c84f1aa6 fix(anthropic.py): compatibility fix 2024-05-11 19:51:29 -07:00
Krrish Dholakia 6d67d6d5ad fix(types/bedrock.py): linting fix 2024-05-11 19:49:46 -07:00
Ishaan Jaff 61a3e5d5a9 fix get healthy deployments 2024-05-11 19:46:35 -07:00
Krrish Dholakia ae0c061b46 fix(anthropic.py): fix version compatibility 2024-05-11 19:46:26 -07:00
Krrish Dholakia b1448cd244 test(test_streaming.py): fix test 2024-05-11 19:44:47 -07:00
Krrish Dholakia 2f3fd3e2f0 fix(anthropic.py): fix linting error 2024-05-11 19:42:14 -07:00
Krrish Dholakia 64650c0279 feat(bedrock_httpx.py): working bedrock command-r sync+async streaming 2024-05-11 19:39:51 -07:00
Ishaan Jaff 04ac352407 test fix - test_async_fallbacks_embeddings 2024-05-11 19:20:24 -07:00
Ishaan Jaff 7930653872 fix - test router fallbacks 2024-05-11 19:13:22 -07:00
Ishaan Jaff 32e445c59d fix - unit tests for router retries 2024-05-11 19:10:33 -07:00
Ishaan Jaff e0d1f96544 test router - fallbacks 2024-05-11 19:08:31 -07:00
Ishaan Jaff 4d648a6d89 fix - _time_to_sleep_before_retry 2024-05-11 19:08:10 -07:00
Ishaan Jaff bfcb640d21 Merge pull request #3590 from BerriAI/litellm_router_retry_logic
[Feat] Proxy + Router - Retry on RateLimitErrors when fallbacks, other deployments exists
2024-05-11 18:21:12 -07:00
Ishaan Jaff c56b44f779 fix failing azure content safety errors 2024-05-11 18:19:00 -07:00
Ishaan Jaff a978326c99 unify sync and async logic for retries 2024-05-11 18:17:04 -07:00
Ishaan Jaff 4e844d7438 test - unit tests for time to sleep when there are rate limit errors 2024-05-11 18:13:28 -07:00
Ishaan Jaff 6e39760779 fix _time_to_sleep_before_retry 2024-05-11 18:05:12 -07:00
Ishaan Jaff 3e6097d9f8 fix _time_to_sleep_before_retry logic 2024-05-11 18:00:02 -07:00