Stub in an initial OpenAI Responses API

Signed-off-by: Ben Browning <bbrownin@redhat.com>
This commit is contained in:
Ben Browning 2025-04-17 14:47:24 -04:00 committed by Ashwin Bharambe
parent c149cf2e0f
commit 70c088af3a
18 changed files with 441 additions and 0 deletions

View file

@ -0,0 +1,5 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.

View file

@ -0,0 +1,90 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
import pytest
from openai import OpenAI
from ..test_cases.test_case import TestCase
@pytest.fixture
def openai_client(client_with_models):
base_url = f"{client_with_models.base_url}/v1/openai/v1"
return OpenAI(base_url=base_url, api_key="bar")
@pytest.mark.parametrize(
"test_case",
[
"openai:responses:non_streaming_01",
"openai:responses:non_streaming_02",
],
)
def test_openai_responses_non_streaming(openai_client, client_with_models, text_model_id, test_case):
tc = TestCase(test_case)
question = tc["question"]
expected = tc["expected"]
response = openai_client.responses.create(
model=text_model_id,
input=question,
stream=False,
)
output_text = response.output_text.lower().strip()
assert len(output_text) > 0
assert expected.lower() in output_text
retrieved_response = openai_client.responses.retrieve(response_id=response.id)
assert retrieved_response.output_text == response.output_text
next_response = openai_client.responses.create(
model=text_model_id, input="Repeat your previous response in all caps.", previous_response_id=response.id
)
next_output_text = next_response.output_text.strip()
assert expected.upper() in next_output_text
@pytest.mark.parametrize(
"test_case",
[
"openai:responses:streaming_01",
"openai:responses:streaming_02",
],
)
def test_openai_responses_streaming(openai_client, client_with_models, text_model_id, test_case):
tc = TestCase(test_case)
question = tc["question"]
expected = tc["expected"]
response = openai_client.responses.create(
model=text_model_id,
input=question,
stream=True,
timeout=120, # Increase timeout to 2 minutes for large conversation history
)
streamed_content = []
response_id = ""
for chunk in response:
response_id = chunk.response.id
streamed_content.append(chunk.response.output_text.strip())
assert len(streamed_content) > 0
assert expected.lower() in "".join(streamed_content).lower()
retrieved_response = openai_client.responses.retrieve(response_id=response_id)
assert retrieved_response.output_text == "".join(streamed_content)
next_response = openai_client.responses.create(
model=text_model_id,
input="Repeat your previous response in all caps.",
previous_response_id=response_id,
stream=True,
)
next_streamed_content = []
for chunk in next_response:
next_streamed_content.append(chunk.response.output_text.strip())
assert expected.upper() in "".join(next_streamed_content)

View file

@ -0,0 +1,26 @@
{
"non_streaming_01": {
"data": {
"question": "Which planet do humans live on?",
"expected": "Earth"
}
},
"non_streaming_02": {
"data": {
"question": "Which planet has rings around it with a name starting with letter S?",
"expected": "Saturn"
}
},
"streaming_01": {
"data": {
"question": "What's the name of the Sun in latin?",
"expected": "Sol"
}
},
"streaming_02": {
"data": {
"question": "What is the name of the US captial?",
"expected": "Washington"
}
}
}

View file

@ -12,6 +12,7 @@ class TestCase:
_apis = [
"inference/chat_completion",
"inference/completion",
"openai/responses",
]
_jsonblob = {}