CVE-2026-41523

vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.0, an assert-based security check in vLLM's activation function loading allows any unauthenticated attacker to achieve arbitrary code execution on the server by publishing a malicious HuggingFace model, when vLLM runs in Python optimized mode (python -O or PYTHONOPTIMIZE=1). This vulnerability is fixed in 0.22.0.
Configurations

Configuration 1 (hide)

cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*

History

07 Jul 2026, 12:16

Type Values Removed Values Added
References
  • () https://access.redhat.com/errata/RHSA-2026:36005 -
  • () https://access.redhat.com/errata/RHSA-2026:36006 -

30 Jun 2026, 03:19

Type Values Removed Values Added
References
  • () https://access.redhat.com/security/cve/CVE-2026-41523 -
  • () https://bugzilla.redhat.com/show_bug.cgi?id=2491582 -
  • () https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-41523.json -

24 Jun 2026, 16:48

Type Values Removed Values Added
First Time Vllm
Vllm vllm
CPE cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*
References () https://github.com/vllm-project/vllm/commit/b3c7ffcab82c2439726f8cb213800f6f38c023d3 - () https://github.com/vllm-project/vllm/commit/b3c7ffcab82c2439726f8cb213800f6f38c023d3 - Patch
References () https://github.com/vllm-project/vllm/security/advisories/GHSA-q8gq-377p-jq3r - () https://github.com/vllm-project/vllm/security/advisories/GHSA-q8gq-377p-jq3r - Exploit, Third Party Advisory
References () https://huntr.com/bounties/dcb05b04-e625-41e7-adbc-bbae0cc2d64c - () https://huntr.com/bounties/dcb05b04-e625-41e7-adbc-bbae0cc2d64c - Third Party Advisory

22 Jun 2026, 23:16

Type Values Removed Values Added
New CVE

Information

Published : 2026-06-22 23:16

Updated : 2026-07-15 02:21


NVD link : CVE-2026-41523

Mitre link : CVE-2026-41523

CVE.ORG link : CVE-2026-41523


JSON object : View

Products Affected

vllm

  • vllm
CWE
CWE-94

Improper Control of Generation of Code ('Code Injection')

CWE-617

Reachable Assertion