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 15, 02:15 UTC
[CI] Fix flaky `test_streaming_response` test (#20913)
Signed-off-by: NickLucche <nlucches@redhat.com>
Nicolò Lucchesi committed
Jul 15, 2025 at 05:15 UTC
80305c1b245eb958c112b3d3d8b8d99c24606637
1 file changed
+4
-2
tests/entrypoints/openai/test_transcription_validation.py
+4
-2
@@ -154,7 +154,8 @@ async def test_streaming_response(winning_call):
154
file=winning_call,
155
language="en",
156
temperature=0.0,
157
- extra_body=dict(stream=True))
157
+ extra_body=dict(stream=True),
158
+ timeout=30)
159
# Reconstruct from chunks and validate
160
async for chunk in res:
161
# just a chunk
@@ -184,7 +185,8 @@ async def test_stream_options(winning_call):
185
temperature=0.0,
186
extra_body=dict(stream=True,
187
stream_include_usage=True,
187
- stream_continuous_usage_stats=True))
188
+ stream_continuous_usage_stats=True),
189
+ timeout=30)
190
final = False
191
continuous = True
192
async for chunk in res: