Aller au contenu
Se connecter
8.2 Élevée

CVE-2026-116178 vLLM from 0.20.0 before 0.30.0 contains a denial of service vulnerability that allows remote attackers to crash the…

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.

Produits et versions touchés

Produit Versions touchées Corrigé en
vllm 0.30 premières → < 0.30.0 0.30.0
vllm 0.20 0.20.0 → < 0.30.0 0.30.0

Autres CVE de vllm

Les 121 CVE de vllm →
CVE-2026-115918 vLLM is an inference and serving engine for large language models 6.5 Moyenne CVE-2026-115917 vLLM is an inference and serving engine for large language models 3.4 Faible CVE-2026-115916 vLLM is an inference and serving engine for large language models 6.5 Moyenne CVE-2026-115915 vLLM is an inference and serving engine for large language models 8.1 Élevée
Suivre cette CVE