CVE-2026-114053
CRITICAL
9,2
Source: [email protected]
Attack Vector: network
Attack Complexity: low
Privileges Required: none
User Interaction: none
Confidentiality: N/A
Integrity: N/A
Availability: N/A
HIGH
8,1
Source: [email protected]
Attack Vector: network
Attack Complexity: high
Privileges Required: none
User Interaction: none
Scope: unchanged
Confidentiality: high
Integrity: high
Availability: high
Description
AI Translation Available
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.
94
Improper Control of Generation of Code ('Code Injection')
DraftCommon Consequences
Security Scopes Affected:
Access Control
Integrity
Confidentiality
Availability
Non-Repudiation
Potential Impacts:
Bypass Protection Mechanism
Gain Privileges Or Assume Identity
Execute Unauthorized Code Or Commands
Hide Activities
Applicable Platforms
Languages:
Interpreted
Technologies:
AI/ML
https://github.com/vllm-project/vllm/commit/570735520942b6428ae98f1eb442c6d88ad…
https://github.com/vllm-project/vllm/security/advisories/GHSA-h3rc-6mm3-gc2m
https://www.vulncheck.com/advisories/vllm-0.7.3-before-0.31.0-rce-via-mm-proces…