@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 1543680691
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 17, 03:00 UTC

[Bugfix] Ensure download_weights_from_hf(..) inside loader is using the revision parameter (#4217)

alexm-nm committed Apr 22, 2024 at 12:10 UTC 15436806912d7ad9371c8bcf6a46857590c107d2
1 file changed +1 -1
vllm/model_executor/model_loader/loader.py
+1 -1
@@ -172,7 +172,7 @@ class DefaultModelLoader(BaseModelLoader):
172 if not is_local:
173 hf_folder = download_weights_from_hf(model_name_or_path,
174 self.load_config.download_dir,
175 - allow_patterns)
175 + allow_patterns, revision)
176 else:
177 hf_folder = model_name_or_path
178