@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / f5a37c6c6c
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, 06:45 UTC

[BugFix] Fix a bug in loading safetensors (#1732)

Woosuk Kwon committed Nov 20, 2023 at 15:51 UTC f5a37c6c6c8204ec42187525ebdb8288cfc4b552
1 file changed +1 -1
vllm/model_executor/weight_utils.py
+1 -1
@@ -240,7 +240,7 @@ def hf_model_weights_iterator(
240 elif use_safetensors:
241 for st_file in hf_weights_files:
242 with safe_open(st_file, framework="pt") as f:
243 - for name in f:
243 + for name in f.keys(): # noqa: SIM118
244 param = f.get_tensor(name)
245 yield name, param
246 else: