Aller au contenu
en Se connecter
6.5 Moyenne

CVE-2026-57173 vLLM is an inference and serving engine for large language models

Description

vLLM is an inference and serving engine for large language models. Prior to 0.24.0, the input_audio handling path for /v1/chat/completions calls AudioMediaIO.load_bytes or AudioMediaIO.load_file without passing VLLM_MAX_AUDIO_DECODE_DURATION_S to the shared audio decoder. An unauthenticated client can therefore submit a small compressed audio input that expands into a very large float32 PCM allocation, bypassing the duration guard already used by /v1/audio/transcriptions and causing an out-of-memory worker crash. Inline data URLs reach this path without being bounded by VLLM_AUDIO_FETCH_TIMEOUT. The issue affects deployments serving an audio-capable model, and authentication changes only the deployment-specific reachability. This issue is fixed in version 0.24.0.

Produits et versions touchés

Produit Versions touchées Corrigé en
vllm 0.24 premières → < 0.24.0 0.24.0

Références

Chronologie

  1. Publiée
  2. Dernière modification

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