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 14, 05:30 UTC
Skip Qwen 1 in CI because remote code is no longer compatible with Transformers (#22536)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Harry Mellor committed
Aug 9, 2025 at 00:20 UTC
41b965575136a72c21927b87a16bd7460b3a3cf8
1 file changed
+2
tests/models/registry.py
+2
@@ -278,6 +278,8 @@ _TEXT_GENERATION_EXAMPLE_MODELS = {
278
transformers_version_reason="vLLM impl inherits PreTrainedModel and clashes with get_input_embeddings", # noqa: E501
279
trust_remote_code=True),
280
"QWenLMHeadModel": _HfExamplesInfo("Qwen/Qwen-7B-Chat",
281
+ max_transformers_version="4.53",
282
+ transformers_version_reason="HF model uses remote code that is not compatible with latest Transformers", # noqa: E501
283
trust_remote_code=True),
284
"Qwen2ForCausalLM": _HfExamplesInfo("Qwen/Qwen2-0.5B-Instruct",
285
extras={"2.5": "Qwen/Qwen2.5-0.5B-Instruct"}), # noqa: E501