CVE-2026-130686

HIGH Status: Awaiting Analysis

CVSS Scores

CVSS v3.x Base Score
7.5
HIGH

Description

vLLM from 0.6.0 before 0.33.0 lacks a request-wide media budget, allowing remote inference clients to exhaust frontend resources by submitting many remote image URLs in one request. Attackers can send a compact request to /v1/messages or /v1/responses that forces downloading and decoding of every image before prompt-length admission, exhausting memory, CPU, and connections.

Published
November 18, 2026 12:17 PM
Last Modified
November 18, 2026 2:17 PM
Source
[email protected]

Weaknesses (CWE)

CWE-400

References

Contact Us

Get the CyboWatch SIEM platform, hosting, and 24×7 SOC analysts. Send an enquiry, or call or email us to get started.

Send an enquiry