tiennm99/litellm
1
0
Fork 0
mirror of https://github.com/tiennm99/litellm.git synced 2026-08-11 22:22:28 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #26525 from BerriAI/litellm_rag_aws_endpoint_cleanup

Browse Source 
[Fix] broaden RAG ingestion credential cleanup to AWS endpoint/identity fields
This commit is contained in:
yuneng-jiang
2026-04-25 15:13:42 -07:00
committed by GitHub
parent 1dee006423 c364c51115
commit b021d5c109
5 changed files with 391 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
litellm/model_prices_and_context_window_backup.json
+163
View File
@@ -4645,6 +4645,169 @@
"supports_vision": true,
"supports_web_search": true
},
"azure/gpt-5.5": {
"cache_read_input_token_cost": 5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1e-06,
"cache_read_input_token_cost_priority": 1e-06,
"cache_read_input_token_cost_above_272k_tokens_priority": 2e-06,
"input_cost_per_token": 5e-06,
"input_cost_per_token_above_272k_tokens": 1e-05,
"input_cost_per_token_priority": 1e-05,
"input_cost_per_token_above_272k_tokens_priority": 2e-05,
"litellm_provider": "azure",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "chat",
"output_cost_per_token": 3e-05,
"output_cost_per_token_above_272k_tokens": 4.5e-05,
"output_cost_per_token_priority": 6e-05,
"output_cost_per_token_above_272k_tokens_priority": 9e-05,
"supported_endpoints": [
"/v1/chat/completions",
"/v1/batch",
"/v1/responses"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true,
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false
},
"azure/gpt-5.5-2026-04-23": {
"cache_read_input_token_cost": 5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1e-06,
"cache_read_input_token_cost_priority": 1e-06,
"cache_read_input_token_cost_above_272k_tokens_priority": 2e-06,
"input_cost_per_token": 5e-06,
"input_cost_per_token_above_272k_tokens": 1e-05,
"input_cost_per_token_priority": 1e-05,
"input_cost_per_token_above_272k_tokens_priority": 2e-05,
"litellm_provider": "azure",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "chat",
"output_cost_per_token": 3e-05,
"output_cost_per_token_above_272k_tokens": 4.5e-05,
"output_cost_per_token_priority": 6e-05,
"output_cost_per_token_above_272k_tokens_priority": 9e-05,
"supported_endpoints": [
"/v1/chat/completions",
"/v1/batch",
"/v1/responses"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true,
"supports_web_search": true
},
"azure/gpt-5.5-pro": {
"cache_read_input_token_cost": 6e-06,
"cache_read_input_token_cost_above_272k_tokens": 1.2e-05,
"input_cost_per_token": 6e-05,
"input_cost_per_token_above_272k_tokens": 0.00012,
"litellm_provider": "azure",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "responses",
"output_cost_per_token": 0.00036,
"output_cost_per_token_above_272k_tokens": 0.00054,
"supported_endpoints": [
"/v1/batch",
"/v1/responses"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": false,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_vision": true,
"supports_web_search": true,
"supports_none_reasoning_effort": false,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_low_reasoning_effort": false
},
"azure/gpt-5.5-pro-2026-04-23": {
"cache_read_input_token_cost": 6e-06,
"cache_read_input_token_cost_above_272k_tokens": 1.2e-05,
"input_cost_per_token": 6e-05,
"input_cost_per_token_above_272k_tokens": 0.00012,
"litellm_provider": "azure",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "responses",
"output_cost_per_token": 0.00036,
"output_cost_per_token_above_272k_tokens": 0.00054,
"supported_endpoints": [
"/v1/batch",
"/v1/responses"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": false,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_vision": true,
"supports_web_search": true
},
"azure/gpt-5.4-mini": {
"cache_read_input_token_cost": 7.5e-08,
"input_cost_per_token": 7.5e-07,
litellm/rag/ingestion/base_ingestion.py
+7 -5
View File
@@ -73,7 +73,7 @@ class BaseRAGIngestion(ABC):
This allows users to specify a credential name in the vector_store config
which will be resolved from litellm.credential_list. When a stored
credential is used, its values take precedence over caller-supplied
equivalents so the api_key / api_base pair stays consistent with the
equivalents so endpoint and identity fields stay consistent with the
credential definition.
"""
from litellm.litellm_core_utils.credential_accessor import CredentialAccessor
@@ -87,11 +87,13 @@ class BaseRAGIngestion(ABC):
return
for key, value in credential_values.items():
self.vector_store_config[key] = value
if (
"api_base" in self.vector_store_config
and "api_base" not in credential_values
for key in (
"api_base",
"aws_sts_endpoint",
"aws_web_identity_token",
):
del self.vector_store_config["api_base"]
if key in self.vector_store_config and key not in credential_values:
del self.vector_store_config[key]
@property
def custom_llm_provider(self) -> str:
model_prices_and_context_window.json
+163
View File
@@ -4659,6 +4659,169 @@
"supports_vision": true,
"supports_web_search": true
},
"azure/gpt-5.5": {
"cache_read_input_token_cost": 5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1e-06,
"cache_read_input_token_cost_priority": 1e-06,
"cache_read_input_token_cost_above_272k_tokens_priority": 2e-06,
"input_cost_per_token": 5e-06,
"input_cost_per_token_above_272k_tokens": 1e-05,
"input_cost_per_token_priority": 1e-05,
"input_cost_per_token_above_272k_tokens_priority": 2e-05,
"litellm_provider": "azure",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "chat",
"output_cost_per_token": 3e-05,
"output_cost_per_token_above_272k_tokens": 4.5e-05,
"output_cost_per_token_priority": 6e-05,
"output_cost_per_token_above_272k_tokens_priority": 9e-05,
"supported_endpoints": [
"/v1/chat/completions",
"/v1/batch",
"/v1/responses"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true,
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false
},
"azure/gpt-5.5-2026-04-23": {
"cache_read_input_token_cost": 5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1e-06,
"cache_read_input_token_cost_priority": 1e-06,
"cache_read_input_token_cost_above_272k_tokens_priority": 2e-06,
"input_cost_per_token": 5e-06,
"input_cost_per_token_above_272k_tokens": 1e-05,
"input_cost_per_token_priority": 1e-05,
"input_cost_per_token_above_272k_tokens_priority": 2e-05,
"litellm_provider": "azure",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "chat",
"output_cost_per_token": 3e-05,
"output_cost_per_token_above_272k_tokens": 4.5e-05,
"output_cost_per_token_priority": 6e-05,
"output_cost_per_token_above_272k_tokens_priority": 9e-05,
"supported_endpoints": [
"/v1/chat/completions",
"/v1/batch",
"/v1/responses"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true,
"supports_web_search": true
},
"azure/gpt-5.5-pro": {
"cache_read_input_token_cost": 6e-06,
"cache_read_input_token_cost_above_272k_tokens": 1.2e-05,
"input_cost_per_token": 6e-05,
"input_cost_per_token_above_272k_tokens": 0.00012,
"litellm_provider": "azure",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "responses",
"output_cost_per_token": 0.00036,
"output_cost_per_token_above_272k_tokens": 0.00054,
"supported_endpoints": [
"/v1/batch",
"/v1/responses"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": false,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_vision": true,
"supports_web_search": true,
"supports_none_reasoning_effort": false,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_low_reasoning_effort": false
},
"azure/gpt-5.5-pro-2026-04-23": {
"cache_read_input_token_cost": 6e-06,
"cache_read_input_token_cost_above_272k_tokens": 1.2e-05,
"input_cost_per_token": 6e-05,
"input_cost_per_token_above_272k_tokens": 0.00012,
"litellm_provider": "azure",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "responses",
"output_cost_per_token": 0.00036,
"output_cost_per_token_above_272k_tokens": 0.00054,
"supported_endpoints": [
"/v1/batch",
"/v1/responses"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": false,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_vision": true,
"supports_web_search": true
},
"azure/gpt-5.4-mini": {
"cache_read_input_token_cost": 7.5e-08,
"input_cost_per_token": 7.5e-07,
tests/test_litellm/litellm_core_utils/llm_cost_calc/test_llm_cost_calc_utils.py
+57
View File
@@ -449,6 +449,63 @@ def test_gpt55_dated_variants_match_base_reasoning_effort_capabilities(
)
@pytest.mark.parametrize(
"model,expected_mode,expected_input,expected_output,expected_cache_read",
[
("azure/gpt-5.5", "chat", 5e-6, 3e-5, 5e-7),
("azure/gpt-5.5-2026-04-23", "chat", 5e-6, 3e-5, 5e-7),
("azure/gpt-5.5-pro", "responses", 6e-5, 3.6e-4, 6e-6),
("azure/gpt-5.5-pro-2026-04-23", "responses", 6e-5, 3.6e-4, 6e-6),
],
)
def test_azure_gpt55_entries_present_with_correct_pricing(
model, expected_mode, expected_input, expected_output, expected_cache_read
):
"""Day-0 Azure entries for GPT-5.5 mirror the OpenAI pricing structure.
Pricing parity with openai/gpt-5.5* (verified against OpenAI's pricing page
on 2026-04-24): $5/$30 input/output per 1M for chat, $60/$360 for pro.
Cache discount is 10% of input.
"""
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
m = litellm.model_cost[model]
assert m["litellm_provider"] == "azure"
assert m["mode"] == expected_mode
assert m["input_cost_per_token"] == expected_input
assert m["output_cost_per_token"] == expected_output
assert m["cache_read_input_token_cost"] == expected_cache_read
# Long-context window inherited from gpt-5.4 / openai gpt-5.5.
assert m["max_input_tokens"] == 1050000
assert m["max_output_tokens"] == 128000
@pytest.mark.parametrize(
"model,expected_none,expected_minimal,expected_xhigh",
[
# Mirror live OpenAI API contract (verified via openai/gpt-5.5* on
# 2026-04-24): chat accepts {none, low, medium, high, xhigh} but NOT
# minimal; pro accepts {medium, high, xhigh} only.
# NOTE: openai/gpt-5.5* entries currently set supports_minimal=true on
# main (pre #26456). Once that PR lands, OpenAI + Azure flags align.
("azure/gpt-5.5", True, False, True),
("azure/gpt-5.5-pro", False, False, True),
],
)
def test_azure_gpt55_reasoning_effort_flags_match_live_openai_api(
model, expected_none, expected_minimal, expected_xhigh
):
"""Azure entries pin reasoning_effort flags to OpenAI's actual API contract."""
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
m = litellm.model_cost[model]
assert m.get("supports_none_reasoning_effort") is expected_none
assert m.get("supports_minimal_reasoning_effort") is expected_minimal
assert m.get("supports_xhigh_reasoning_effort") is expected_xhigh
def test_generic_cost_per_token_anthropic_prompt_caching():
model = "claude-sonnet-4@20250514"
usage = Usage(
tests/test_litellm/test_utils.py
+1
View File
@@ -769,6 +769,7 @@ def test_aaamodel_prices_and_context_window_json_is_valid():
"uses_embed_content": {"type": "boolean"},
"supports_reasoning": {"type": "boolean"},
"supports_minimal_reasoning_effort": {"type": "boolean"},
"supports_low_reasoning_effort": {"type": "boolean"},
"supports_none_reasoning_effort": {"type": "boolean"},
"supports_xhigh_reasoning_effort": {"type": "boolean"},
"supports_max_reasoning_effort": {"type": "boolean"},