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 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.

CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H

可信数据的反序列化

vLLM 代码问题漏洞

vLLM是vLLM开源的一个适用于 LLM 的高吞吐量和内存高效推理和服务引擎。 vLLM v0.7.0之前版本存在代码问题漏洞，该漏洞源于当torch.load加载恶意pickle数据时，会在unpickling期间执行任意代码。

N/A

N/A