Vulnerability in vllm-project vllm

[ Description ]Official description

vLLM from 0.20.0 before 0.30.0 contains a denial of service vulnerability that allows remote attackers to crash the engine by sending inconsistent multimodal metadata to the disaggregated /inference/v1/generate endpoint. Attackers can replay a valid multimodal item with repeated or overlapping mm_placeholders ranges or mismatched kwargs_data and mm_hashes lists, producing a negative text span in Qwen2.5-VL position construction that terminates the engine.

[ Fixes ]Fixes and patches

• Patch on github.comVendor patch or bulletin · NVD · Reference tagged 'Patch' by NVD

[ Advisories ]Government CERT advisories

• BSI CERT-Bund8 Oct 2026WID-SEC-2026-3828

  vllm: Multiple vulnerabilities Translated from German

  Risk mediumCVE-2026-116176CVE-2026-116180CVE-2026-116177+4 more CVEs

[ References ]References

• github.com/vllm-project/vllm/security/advisories/GHSA-34qw-q2h2-jp3mPatchVendor Advisory
• www.vulncheck.com/advisories/vllm-0.20.0-before-0.30.0-dos-via-inference-v1-generate-mm…Third Party Advisory