CVE-2026-126397
Published:November 13, 2026
Updated:November 16, 2026
vLLM from 0.10.1 before 0.31.0 forwards a request-controlled mm_processor_kwargs.max_patches value to the Hugging Face processor for Cohere2 Vision models without a server-side upper bound, causing quadratic CPU work during multimodal preprocessing. Remote clients can send small image chat-completion requests with large, varying max_patches values to bypass the LRU cache, exhaust preprocessing workers, and deny service to other clients.
Affected Packages
vllm (CONDA):
Affected version(s) >=0.10.1 <0.31.0Fix Suggestion:
Update to version 0.31.0https://github.com/vllm-project/vllm.git (GITHUB):
Affected version(s) >=v0.10.1rc1 <v0.31.0Fix Suggestion:
Update to version v0.31.0vllm (PYTHON):
Affected version(s) >=0.10.1 <0.31.0Fix Suggestion:
Update to version 0.31.0Related Resources (5)
Do you need more information?
Contact UsCVSS v4
Base Score:
8.7
Attack Vector
NETWORK
Attack Complexity
LOW
Attack Requirements
NONE
Privileges Required
NONE
User Interaction
NONE
Vulnerable System Confidentiality
NONE
Vulnerable System Integrity
NONE
Vulnerable System Availability
HIGH
Subsequent System Confidentiality
NONE
Subsequent System Integrity
NONE
Subsequent System Availability
NONE
CVSS v3
Base Score:
7.5
Attack Vector
NETWORK
Attack Complexity
LOW
Privileges Required
NONE
User Interaction
NONE
Scope
UNCHANGED
Confidentiality
NONE
Integrity
NONE
Availability
HIGH
Weakness Type (CWE)
Uncontrolled Resource Consumption
EPSS
Base Score:
0.46