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, 16:00 UTC
[Misc] improve gguf check (#15974)
Signed-off-by: reidliu41 <reid201711@gmail.com> Co-authored-by: reidliu41 <reid201711@gmail.com>
Reid committed
Apr 4, 2025 at 09:33 UTC
86cbd2eee97a98df59c531c34d2aeff5a2b5765d
1 file changed
+8
-3
vllm/transformers_utils/utils.py
+8
-3
@@ -23,9 +23,14 @@ def check_gguf_file(model: Union[str, PathLike]) -> bool:
23
elif model.suffix == ".gguf":
24
return True
25
26
- with open(model, "rb") as f:
27
- header = f.read(4)
28
- return header == b"GGUF"
26
+ try:
27
+ with model.open("rb") as f:
28
+ header = f.read(4)
29
+
30
+ return header == b"GGUF"
31
+ except Exception as e:
32
+ logger.debug("Error reading file %s: %s", model, e)
33
+ return False
34
35
36
def modelscope_list_repo_files(