Description
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.
Metrics
CVSS 3.1 8.1/10
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:HWeakness enumeration
Affected software
| Vendor | Product | Versions |
|---|---|---|
| vllm-project | vllm | >= 0.7.3, < 0.31.0 |
References
- https://github.com/vllm-project/vllm/commit/570735520942b6428ae98f1eb442c6d88adae55ePatch
- https://github.com/vllm-project/vllm/security/advisories/GHSA-h3rc-6mm3-gc2mVendor Advisory
- https://www.vulncheck.com/advisories/vllm-0.7.3-before-0.31.0-rce-via-mm-processor-kwargs-code-revisionThird Party Advisory
Timeline
- Published
- Oct 21, 2026
- Last modified
- Nov 3, 2026
- NVD status
- Analyzed