@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / ab6f981671
Fork 0

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, 09:30 UTC

[CI][Bugfix] Skip chameleon for transformers 4.46.1 (#9808)

Michael Goin committed Oct 29, 2024 at 14:12 UTC ab6f981671c4e5035575f5e5ef6172f4df52e121
1 file changed +1 -1
tests/models/decoder_only/vision_language/test_broadcast.py
+1 -1
@@ -24,7 +24,7 @@ def test_models(hf_runner, vllm_runner, image_assets,
24 elif model.startswith("llava-hf/llava-v1.6"):
25 from .test_llava_next import models, run_test # type: ignore[no-redef]
26 elif model.startswith("facebook/chameleon"):
27 - if transformers.__version__.startswith("4.46.0"):
27 + if transformers.__version__.startswith("4.46"):
28 pytest.skip("Model broken in HF, "
29 "see huggingface/transformers#34379")
30 from .test_chameleon import models, run_test # type: ignore[no-redef]