CVE-2025-24357 — vLLM allows a malicious model RCE by torch.load in hf_model_weights_iterator

vLLM is a library for LLM inference and serving. vllm/model_executor/weight_utils.py implements hf_model_weights_iterator to load the model checkpoint, which is downloaded from …

Published: 2025-01-27 · CVSS: 7.5 · CVSS 3.1

vLLM is a library for LLM inference and serving. vllm/model_executor/weight_utils.py implements hf_model_weights_iterator to load the model checkpoint, which is downloaded from huggingface. It uses the torch.load function and the weights_only parameter defaults to False. When torch.load loads malicious pickle data, it will execute arbitrary code during unpickling. This vulnerability is fixed in v0.7.0.

Original CVE JSON · STIX 2.1

Source: CVE Program · Revision 8b16575c39f1e1810c5e863010da073d2d3285e7