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 6, 13:15 UTC
[ROCm][CI] Add ROCM_EXTRA_ARGS to audio_in_video test server fixture (#37349)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
Andreas Karatzas committed
Mar 17, 2026 at 23:55 UTC
8b6325758cce5f9c36d38f2462edbd368b97a07c
1 file changed
+2
-1
tests/entrypoints/openai/chat_completion/test_audio_in_video.py
+2
-1
@@ -9,7 +9,7 @@ import pytest
9
import pytest_asyncio
10
11
from tests.conftest import VideoTestAssets
12
-from tests.utils import RemoteOpenAIServer
12
+from tests.utils import ROCM_EXTRA_ARGS, RemoteOpenAIServer
13
14
MODEL_NAME = "Qwen/Qwen2.5-Omni-3B"
15
@@ -22,6 +22,7 @@ def server():
22
"--enforce-eager",
23
"--limit-mm-per-prompt",
24
json.dumps({"audio": 3, "video": 3}),
25
+ *ROCM_EXTRA_ARGS,
26
]
27
28
with RemoteOpenAIServer(