CyberRota Analysis
AI-GeneratedVersions 0.12.1 through 0.12.2 of LMDeploy are vulnerable to a code injection flaw that allows attackers to execute arbitrary Python code by manipulating the `quantization_config.quant_dtype` parameter in a malicious HuggingFace model. This vulnerability poses a significant risk to users who load models using LMDeploy, as it can lead to unauthorized code execution in their environment. Organizations utilizing LMDeploy should prioritize upgrading to version 0.12.3 or later to mitigate this high-severity risk.
Public Exploit Signal
A public exploit, PoC, GitHub repository or Metasploit reference was detected for this CVE.
Note: these links are listed for security research and verification purposes only.
Original NVD Description
LMDeploy is a toolkit for compressing, deploying, and serving large language models. Versions 012.1 through 0.12.2 contain a code injection vulnerability in `lmdeploy/pytorch/config.py` line 620 that allows an attacker to execute arbitrary Python code by publishing a malicious HuggingFace model with a crafted `quantization_config.quant_dtype` value. When a user loads the model with lmdeploy, the `quant_dtype` is passed to `eval(f'torch.{quant_dtype}')` without any validation. Version 0.12.3 contains a patch.