CVE-2026-126399

CVSS 7.1 (high) · published 2026-11-13

vLLM from 0.19.0 before 0.31.0 contains a resource exhaustion vulnerability that allows inference clients to bypass the server frame limit by setting media_io_kwargs.video.num_frames in chat completion requests. Attackers can supply num_frames=-1 or a large value with many base64 JPEG frames, forcing VideoMediaIO to decode all frames and np.stack them, causing memory exhaustion.

Weakness: CWE-770

Source: NVD CVE-2026-126399

References

• https://github.com/vllm-project/vllm/security/advisories/GHSA-vxqj-p4gw-9h4c
• https://www.vulncheck.com/advisories/vllm-0.19.0-before-0.31.0-resource-exhaustion-via-request-level-num-frames

Check your software

Connect an AI agent to the FixLane MCP server or ask the Telegram bot to check what you have installed against incidents with a verified fixed version.