GHSA-3hmm-rh5q-gwwr: LMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loading
Source: github.com
lmdeploy <= latest contains 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. …
Signalsas the pipeline stamped them
High High-risk vulnerability — CVSS 8.8
CVEKEV and EPSS where known
Get the next one first
Every story goes to Telegram and RSS the moment it is published, and the whole wire is a keyless JSON API.