phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-25 18:54:30 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

[Feat] Expose Responses API on LiteLLM UI Test Key Page (#10166)

Browse source 
* add /responses API on UI

* add makeOpenAIResponsesRequest

* add makeOpenAIResponsesRequest

* fix add responses API on UI

* fix endpoint selector

* responses API render chunks on litellm chat ui

* fixes to streaming iterator

* fix render responses completed events

* fixes for MockResponsesAPIStreamingIterator

* transform_responses_api_request_to_chat_completion_request

* fix for responses API

* test_basic_openai_responses_api_streaming

* fix base responses api tests
This commit is contained in:
Ishaan Jaff 2025-04-19 13:18:54 -07:00 • committed by GitHub
parent 518a4900f6
commit 17f55e9937
8 changed files with 332 additions and 52 deletions
Download patch file Download diff file Expand all files Collapse all files

38
tests/llm_responses_api_testing/base_responses_api.py
View file

@ -133,11 +133,13 @@ class BaseResponsesAPITest(ABC):
validate_responses_api_response(response, final_chunk=True)
@pytest.mark.parametrize("sync_mode", [True])
@pytest.mark.parametrize("sync_mode", [True, False])
@pytest.mark.asyncio
async def test_basic_openai_responses_api_streaming(self, sync_mode):
litellm._turn_on_debug()
base_completion_call_args = self.get_base_completion_call_args()
collected_content_string = ""
response_completed_event = None
if sync_mode:
response = litellm.responses(
input="Basic ping",
@ -146,6 +148,10 @@ class BaseResponsesAPITest(ABC):
)
for event in response:
print("litellm response=", json.dumps(event, indent=4, default=str))
if event.type == "response.output_text.delta":
collected_content_string += event.delta
elif event.type == "response.completed":
response_completed_event = event
else:
response = await litellm.aresponses(
input="Basic ping",
@ -154,5 +160,35 @@ class BaseResponsesAPITest(ABC):
)
async for event in response:
print("litellm response=", json.dumps(event, indent=4, default=str))
if event.type == "response.output_text.delta":
collected_content_string += event.delta
elif event.type == "response.completed":
response_completed_event = event
# assert the delta chunks content had len(collected_content_string) > 0
# this content is typically rendered on chat ui's
assert len(collected_content_string) > 0
# assert the response completed event is not None
assert response_completed_event is not None
# assert the response completed event has a response
assert response_completed_event.response is not None
# assert the response completed event includes the usage
assert response_completed_event.response.usage is not None
# basic test assert the usage seems reasonable
print("response_completed_event.response.usage=", response_completed_event.response.usage)
assert response_completed_event.response.usage.input_tokens > 0 and response_completed_event.response.usage.input_tokens < 100
assert response_completed_event.response.usage.output_tokens > 0 and response_completed_event.response.usage.output_tokens < 1000
assert response_completed_event.response.usage.total_tokens > 0 and response_completed_event.response.usage.total_tokens < 1000
# total tokens should be the sum of input and output tokens
assert response_completed_event.response.usage.total_tokens == response_completed_event.response.usage.input_tokens + response_completed_event.response.usage.output_tokens