← Back to CVE List
CVE-2026-114053NVD
Vulnerability Summary
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.
CVSS v4.0 Base Metrics — Score 9.2 (CRITICAL)
Attack VectorNetwork
Attack ComplexityLow
Attack RequirementsPresent
Privileges RequiredNone
User InteractionNone
Confidentiality (Vulnerable System)High
Integrity (Vulnerable System)High
Availability (Vulnerable System)High
Confidentiality (Subsequent System)None
Integrity (Subsequent System)None
Availability (Subsequent System)None
CVSS v3.1 Base Metrics — Score 8.1 (HIGH)
Attack VectorNetwork
Attack ComplexityHigh
Privileges RequiredNone
User InteractionNone
ScopeUnchanged
ConfidentialityHigh
IntegrityHigh
AvailabilityHigh
Affected & Patched Versions
- vllm-project vllm >= 0.7.3 and < 0.31.0
- vllm-project vllm 0.31.0