vLLM: Request-controlled mm_processor_kwargs.code_revision allows remote code execution
CVE-2026-114053 · GHSA-h3rc-6mm3-gc2m · published 2026-10-27
- Severity
- High
- CVSS score
- 8.1
- Chance of exploitation (EPSS)
- 0.46%
- Fixed in
- 0.31.0
- Exploitation No known exploitation
- Automatable No
- Technical impact Total
- CISA SSVC
An untrusted multimodal request can set mmprocessorkwargs.coderevision. When vLLM serves a model whose Hugging Face processor uses dynamic automap code and the operator enabled --trust-remote-code, this request value reaches AutoProcessor.from_pretrained and selects the Python revision imported by Transformers. This lets a client replace the operator-reviewed processor module with code from another revision of the same model repository. The operator-selected model/data revision remains pinned; only the executed processor code changes. Importing a Python module is arbitrary code execution in the vLLM API/renderer process.
Upgrade vllm to 0.31.0 or later. If you start it from an MCP config with npx or uvx, pin that version so an old cached copy is not used.
1. Separate processor-construction arguments from untrusted processor-call arguments. An allowlist of supported request-time preprocessing controls is safest. 2. Reject coderevision and other loader/source controls in request mmprocessorkwargs, and pass modelconfig.code_revision explicitly as the only permitted processor-code revision. 3. Add a regression test proving a request cannot change the loaded module revision while operator pins remain effective. 4. Protect /tokenize with the configured authentication policy as defense in depth. Expected safe behavior is either a 4xx response for reserved request keys or silently ignoring them before processor construction.
| Package | Registry | Affected | Fixed in |
|---|---|---|---|
| vllm | PyPI | >= 0.7.3, < 0.31.0 | 0.31.0 |
- CriticalUnauthenticated Remote Code Execution in Langflow via Public Flow Build EndpointlangflowExploited
- CriticalRay is vulnerable to Critical RCE via Safari & Firefox Browsers through DNS Rebinding AttackrayExploited
- CriticalLangflow Unauth RCElangflow, langflow-baseExploited
- HighLiteLLM: Authenticated command execution via MCP stdio test endpointslitellmExploited
- CriticalMLflow vulnerable to remote code execution on unauthenticated servers via third-party scorer deserialization importing caller-placed artifacmlflow
- CriticalAgno: Arbitrary code execution via `eval()` on `field_type` in `FunctionCall`agno
- CriticalvLLM: OpenAI auth bypassvllm
- CriticalvLLM has RCE In Video Processingvllm
- CriticalvLLM Allows Remote Code Execution via PyNcclPipe Communication Servicevllm
- CriticalvLLM Vulnerable to Remote Code Execution via Mooncake Integrationvllm
- CriticalCVE-2025-24357 Malicious model remote code execution fix bypass with PyTorch < 2.6.0vllm
- CriticalvLLM Deserialization of Untrusted Data vulnerabilityvllm
The impact, patch and workaround text comes from the advisory, shortened (GitHub Advisory Database, CC BY 4.0). Proof-of-concept and reproduction sections and code are left out; the original advisory has them.