GHSA-q8gq-377p-jq3r: vllm (CVE-2026-41523)

Package	Ecosystem	Affected	Fixed in
vllm	PyPI	<0.22.0	0.22.0

Machine-written summary

Summary written by Claude Haiku from the source advisory. Not reviewed by a person.

vLLM versions before 0.22.0 are affected. When vLLM runs in Python optimized mode (python -O or PYTHONOPTIMIZE=1), an assert-based check that restricts which activation functions can be loaded from a HuggingFace model's config.json is removed, so a malicious model publisher can cause arbitrary code execution on a victim's server when the victim loads a crafted cross-encoder model. Upgrade to vLLM 0.22.0 or later. The source gives no workaround beyond avoiding optimized mode and loading untrusted models.

Written 4 Nov 2026.

Not analysed yet

We haven’t published our analysis of this Advisory yet. Once a person has reviewed it, this page will say what’s vulnerable, what has to be true for your app to be affected, and how to check.

Severity and exploitation

Score	Rating	CVSS	Vector	Source
7.5	High	3.1	CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H	CNA (GitHub_M)

CISA KEV Not listed.
EPSS 0.0091, the 60th percentile, on 14 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.

Timeline

31 Oct 2026, 14:29 UTC First seen by agentprime
31 Oct 2026, 14:29 UTC Confirmed by OSV

Sources

Facts on this page come from the GitHub Advisory Database (CC-BY 4.0), OSV and the other sources listed, as published. We don’t reproduce full advisory descriptions, which can contain exploit code.