CVE-2026-114053
vLLM from 0.7.3 before 0.31.0 forwards the request-supplied mm_processor_kwargs.code_revision to Transformers AutoProcessor.from_pretrained, letting clients select which revision of a model's dynamic processor code is imported when trust_remote_code is enabled. An attacker controlling a revision in the served model repository can send a multimodal request to the unauthenticated /tokenize endpoint to execute arbitrary Python in the vLLM API process.
- Score
- CVSS 4.0 9.2 Critical
- Status
- Undergoing Analysis - disclosure@vulncheck.com
- Published
- 2026-10-21
- Last modified
- 2026-10-27
- Weaknesses
- CWE-94
Affected products
References
- https://github.com/vllm-project/vllm/security/advisories/GHSA-h3rc-6mm3-gc2m
- https://github.com/vllm-project/vllm/commit/570735520942b6428ae98f1eb442c6d88adae55e
- https://www.vulncheck.com/advisories/vllm-0.7.3-before-0.31.0-rce-via-mm-processor-kwargs-code-revision
