tiennm99/litellm
1
0
Fork 0
mirror of https://github.com/tiennm99/litellm.git synced 2026-07-11 07:05:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

[Fix] Support service_tier in chat completion (#15693)

Browse Source 
* Support service_tier

* fix test
This commit is contained in:
katsuhiro muto
2025-10-19 05:55:54 +09:00
committed by GitHub
parent c1355e92dc
commit d5e686b3e8
6 changed files with 102 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
litellm/constants.py
+2
View File
@@ -399,6 +399,7 @@ OPENAI_CHAT_COMPLETION_PARAMS = [
"extra_headers",
"thinking",
"web_search_options",
"service_tier",
]
OPENAI_TRANSCRIPTION_PARAMS = [
@@ -453,6 +454,7 @@ DEFAULT_CHAT_COMPLETION_PARAM_VALUES = {
"reasoning_effort": None,
"thinking": None,
"web_search_options": None,
"service_tier": None,
"safety_identifier": None,
}
litellm/llms/openai/chat/gpt_transformation.py
+1
View File
@@ -158,6 +158,7 @@ class OpenAIGPTConfig(BaseLLMModelInfo, BaseConfig):
"parallel_tool_calls",
"audio",
"web_search_options",
"service_tier",
"safety_identifier",
] # works across all models
litellm/main.py
+4
View File
@@ -380,6 +380,7 @@ async def acompletion(
Literal["none", "minimal", "low", "medium", "high", "default"]
] = None,
safety_identifier: Optional[str] = None,
service_tier: Optional[str] = None,
# set api_base, api_version, api_key
base_url: Optional[str] = None,
api_version: Optional[str] = None,
@@ -529,6 +530,7 @@ async def acompletion(
"model_list": model_list,
"reasoning_effort": reasoning_effort,
"safety_identifier": safety_identifier,
"service_tier": service_tier,
"extra_headers": extra_headers,
"acompletion": True, # assuming this is a required parameter
"thinking": thinking,
@@ -951,6 +953,7 @@ def completion( # type: ignore # noqa: PLR0915
deployment_id=None,
extra_headers: Optional[dict] = None,
safety_identifier: Optional[str] = None,
service_tier: Optional[str] = None,
# soon to be deprecated params by OpenAI
functions: Optional[List] = None,
function_call: Optional[str] = None,
@@ -1293,6 +1296,7 @@ def completion( # type: ignore # noqa: PLR0915
"thinking": thinking,
"web_search_options": web_search_options,
"safety_identifier": safety_identifier,
"service_tier": service_tier,
"allowed_openai_params": kwargs.get("allowed_openai_params"),
}
optional_params = get_optional_params(
litellm/model_prices_and_context_window_backup.json
+18
View File
@@ -12520,6 +12520,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4.1-2025-04-14": {
@@ -12553,6 +12554,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4.1-mini": {
@@ -12589,6 +12591,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4.1-mini-2025-04-14": {
@@ -12622,6 +12625,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4.1-nano": {
@@ -12658,6 +12662,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4.1-nano-2025-04-14": {
@@ -12691,6 +12696,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4.5-preview": {
@@ -12755,6 +12761,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4o-2024-05-13": {
@@ -12795,6 +12802,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4o-2024-11-20": {
@@ -12815,6 +12823,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4o-audio-preview": {
@@ -12906,6 +12915,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4o-mini-2024-07-18": {
@@ -12931,6 +12941,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4o-mini-audio-preview": {
@@ -13248,6 +13259,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-5-pro": {
@@ -13485,6 +13497,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-5-mini-2025-08-07": {
@@ -13523,6 +13536,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-5-nano": {
@@ -16543,6 +16557,7 @@
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"o3-2025-04-16": {
@@ -16574,6 +16589,7 @@
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"o3-deep-research": {
@@ -16758,6 +16774,7 @@
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"o4-mini-2025-04-16": {
@@ -16776,6 +16793,7 @@
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"o4-mini-deep-research": {
model_prices_and_context_window.json
+18
View File
@@ -12520,6 +12520,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4.1-2025-04-14": {
@@ -12553,6 +12554,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4.1-mini": {
@@ -12589,6 +12591,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4.1-mini-2025-04-14": {
@@ -12622,6 +12625,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4.1-nano": {
@@ -12658,6 +12662,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4.1-nano-2025-04-14": {
@@ -12691,6 +12696,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4.5-preview": {
@@ -12755,6 +12761,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4o-2024-05-13": {
@@ -12795,6 +12802,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4o-2024-11-20": {
@@ -12815,6 +12823,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4o-audio-preview": {
@@ -12906,6 +12915,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4o-mini-2024-07-18": {
@@ -12931,6 +12941,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-4o-mini-audio-preview": {
@@ -13248,6 +13259,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-5-pro": {
@@ -13485,6 +13497,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-5-mini-2025-08-07": {
@@ -13523,6 +13536,7 @@
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"gpt-5-nano": {
@@ -16543,6 +16557,7 @@
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"o3-2025-04-16": {
@@ -16574,6 +16589,7 @@
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"o3-deep-research": {
@@ -16758,6 +16774,7 @@
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"o4-mini-2025-04-16": {
@@ -16776,6 +16793,7 @@
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true
},
"o4-mini-deep-research": {
tests/llm_translation/test_openai.py
+59
View File
@@ -728,6 +728,65 @@ def test_openai_safety_identifier_parameter_sync():
assert request_body["safety_identifier"] == "user_code_123456"
@pytest.mark.asyncio
async def test_openai_service_tier_parameter():
"""Test that service_tier parameter is correctly passed to the OpenAI API."""
from openai import AsyncOpenAI
litellm.set_verbose = True
client = AsyncOpenAI(api_key="fake-api-key")
with patch.object(
client.chat.completions.with_raw_response, "create"
) as mock_client:
try:
await litellm.acompletion(
model="openai/gpt-4o",
messages=[{"role": "user", "content": "Hello, how are you?"}],
service_tier="priority",
client=client,
)
except Exception as e:
print(f"Error: {e}")
mock_client.assert_called_once()
request_body = mock_client.call_args.kwargs
# Verify the request contains the service_tier parameter
assert "service_tier" in request_body, "service_tier should be in request body"
# Verify service_tier is correctly sent to the API
assert request_body["service_tier"] == "priority", "service_tier should be 'priority'"
def test_openai_service_tier_parameter_sync():
"""Test that service_tier parameter is correctly passed to the OpenAI API."""
from openai import OpenAI
litellm.set_verbose = True
client = OpenAI(api_key="fake-api-key")
with patch.object(
client.chat.completions.with_raw_response, "create"
) as mock_client:
try:
litellm.completion(
model="openai/gpt-4o",
messages=[{"role": "user", "content": "Hello, how are you?"}],
service_tier="priority",
client=client,
)
except Exception as e:
print(f"Error: {e}")
mock_client.assert_called_once()
request_body = mock_client.call_args.kwargs
# Verify the request contains the service_tier parameter
assert "service_tier" in request_body, "service_tier should be in request body"
# Verify service_tier is correctly sent to the API
assert request_body["service_tier"] == "priority", "service_tier should be 'priority'"
def test_gpt_5_reasoning_streaming():
litellm._turn_on_debug()
response = litellm.completion(