A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 16, 04:30 UTC
[CI] Fix flaky entrypoint tests (#11403)
Signed-off-by: Roger Wang <ywang@roblox.com>
Roger Wang committed
Dec 21, 2024 at 21:08 UTC
29c748930e0d35a98351a8cf8a093fba4b758114
3 files changed
+9
tests/entrypoints/openai/test_audio.py
+3
@@ -74,6 +74,7 @@ async def test_single_chat_session_audio(client: openai.AsyncOpenAI,
74
messages=messages,
75
max_completion_tokens=10,
76
logprobs=True,
77
+ temperature=0.0,
78
top_logprobs=5)
79
assert len(chat_completion.choices) == 1
80
@@ -130,6 +131,7 @@ async def test_single_chat_session_audio_base64encoded(
131
messages=messages,
132
max_completion_tokens=10,
133
logprobs=True,
134
+ temperature=0.0,
135
top_logprobs=5)
136
assert len(chat_completion.choices) == 1
137
@@ -150,6 +152,7 @@ async def test_single_chat_session_audio_base64encoded(
152
model=model_name,
153
messages=messages,
154
max_completion_tokens=10,
155
+ temperature=0.0,
156
)
157
message = chat_completion.choices[0].message
158
assert message.content is not None and len(message.content) >= 0
tests/entrypoints/openai/test_video.py
+3
@@ -82,6 +82,7 @@ async def test_single_chat_session_video(client: openai.AsyncOpenAI,
82
messages=messages,
83
max_completion_tokens=10,
84
logprobs=True,
85
+ temperature=0.0,
86
top_logprobs=5)
87
assert len(chat_completion.choices) == 1
88
@@ -174,6 +175,7 @@ async def test_single_chat_session_video_base64encoded(
175
messages=messages,
176
max_completion_tokens=10,
177
logprobs=True,
178
+ temperature=0.0,
179
top_logprobs=5)
180
assert len(chat_completion.choices) == 1
181
@@ -194,6 +196,7 @@ async def test_single_chat_session_video_base64encoded(
196
model=model_name,
197
messages=messages,
198
max_completion_tokens=10,
199
+ temperature=0.0,
200
)
201
message = chat_completion.choices[0].message
202
assert message.content is not None and len(message.content) >= 0
tests/entrypoints/openai/test_vision.py
+3
@@ -83,6 +83,7 @@ async def test_single_chat_session_image(client: openai.AsyncOpenAI,
83
messages=messages,
84
max_completion_tokens=10,
85
logprobs=True,
86
+ temperature=0.0,
87
top_logprobs=5)
88
assert len(chat_completion.choices) == 1
89
@@ -175,6 +176,7 @@ async def test_single_chat_session_image_base64encoded(
176
messages=messages,
177
max_completion_tokens=10,
178
logprobs=True,
179
+ temperature=0.0,
180
top_logprobs=5)
181
assert len(chat_completion.choices) == 1
182
@@ -195,6 +197,7 @@ async def test_single_chat_session_image_base64encoded(
197
model=model_name,
198
messages=messages,
199
max_completion_tokens=10,
200
+ temperature=0.0,
201
)
202
message = chat_completion.choices[0].message
203
assert message.content is not None and len(message.content) >= 0