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 10, 12:45 UTC
[CI] Skip Qwen-VL in multimodal processing tests due to flaky external dependency (#31932)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
Andreas Karatzas committed
Jan 7, 2026 at 20:58 UTC
8dd2419fa96488bc862ad60f7fc7ffffa39728e7
1 file changed
+5
tests/models/multimodal/processing/test_common.py
+5
@@ -404,6 +404,11 @@ def test_processing_correctness(
404
pytest.skip("Fix later")
405
if model_id == "jinaai/jina-reranker-m0":
406
pytest.skip("Fix later")
407
+ if model_id in {"Qwen/Qwen-VL", "Qwen/Qwen-VL-Chat"}:
408
+ pytest.skip(
409
+ "Qwen-VL tokenizer requires downloading a font file from "
410
+ "servers that often refuse connections in CI"
411
+ )
412
413
_test_processing_correctness(
414
model_id,