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, 17:15 UTC
[Bugfix][V1] Avoid importing PreTrainedModel (#15366)
Signed-off-by: Hollow Man <hollowman@opensuse.org>
ℍ𝕠𝕝𝕝𝕠𝕨 𝕄𝕒𝕟 committed
Mar 24, 2025 at 12:33 UTC
948ab03e7ed72d3715f8b7b6398d1d5bb0f9a296
1 file changed
+1
-1
vllm/model_executor/model_loader/utils.py
+1
-1
@@ -32,7 +32,7 @@ def set_default_torch_dtype(dtype: torch.dtype):
32
33
def is_transformers_impl_compatible(
34
arch: str,
35
- module: Optional[transformers.PreTrainedModel] = None) -> bool:
35
+ module: Optional["transformers.PreTrainedModel"] = None) -> bool:
36
mod = module or getattr(transformers, arch, None)
37
if mod is None:
38
return False