Advisory

PYSEC-2025-58

Also known as: CVE-2025-24357, GHSA-rh4j-5rhw-hr54

Summary

The record has no summary.

Details

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.

Severity

High Basis: CVSS v3 score 8.8, vector CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Dates

Published
2025-01-27 18:15 UTC
Modified
2025-06-27 21:56 UTC

Affected packages and fixes

vllm

PyPI · All systems

Affected

  • introduced 0, fixed 0.7.0 (ecosystem)

Fixed versions

0.7.0

Fix

  • Fixed version 0.7.0
    pip install --upgrade vllm==0.7.0

Fix commands are the export's own generic commands. Ojuju never runs them. Check them against your setup first.

Check your own computer: guard packages check

View this record on OSV.dev

Source

Source: PyPA Advisory Database · CC BY 4.0

Sources & licences