6.5Media

CVE-2026-34755vLLM Affected by Denial of Service via Unbounded Frame Count in video/jpeg Base64 Processing

vllm-project vllm; Red Hat Red Hat AI Inference Server 3.2; Red Hat Red Hat AI Inference Server 3.4; Red Hat Red Hat Enterprise Linux AI 3.4 Publicada: (hace 7 meses)

Descripción

vLLM is an inference and serving engine for large language models (LLMs). From 0.7.0 to before 0.19.0, the VideoMediaIO.load_base64() method at vllm/multimodal/media/video.py splits video/jpeg data URLs by comma to extract individual JPEG frames, but does not enforce a frame count limit. The num_frames parameter (default: 32), which is enforced by the load_bytes() code path, is completely bypassed in the video/jpeg base64 path. An attacker can send a single API request containing thousands of comma-separated base64-encoded JPEG frames, causing the server to decode all frames into memory and crash with OOM. This vulnerability is fixed in 0.19.0.

Debilidad (CWE): CWE-770

Datos: CVE.org, NVD, CISA KEV, FIRST EPSS, Ubuntu. Se publican tal como los indican las fuentes, en inglés.

Productos y versiones afectados

FabricanteProductoVersiones
ai inference server 3.2 —
ai inference server 3.4 —
enterprise linux ai 3.4 —
Red Hat Red Hat AI Inference Server all versions
Red Hat Red Hat AI Inference Server 3.2 all versions
Red Hat Red Hat AI Inference Server 3.4 all versions
Red Hat Red Hat Enterprise Linux AI 3.4 all versions
Red Hat Red Hat OpenShift AI (RHOAI) all versions
vllm-project vllm >= 0.7.0, < 0.19.0

Referencias y parches

Coincidencias CPE (NVD) (1)

  • cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:* ≥ 0.7.0 < 0.19.0