Skip to content
CORVASEC — Knowledge Defends
Vulnerability Explorer›CVE-2026-69147
VULNERABILITY

CVE-2026-69147

Common Vulnerabilities and Exposures record

MEDIUM

Summary

vLLM is an inference and serving engine for large language models. Prior to 0.28.0, request bodies for Chat Completions and Responses can set media_io_kwargs.video.video_backend to pynvvideocodec, and MediaConnector.fetch_video forwards that choice to VideoMediaIO even when startup configuration selected a software decoder. The engine's _reserve_mm_ipc_gpu_memory logic budgets decoder memory only from static configuration, so the request-selected VIDEO_LOADER_REGISTRY backend can create a CUDA context, decoder surfaces, and decoded-frame allocations that were not removed from the engine's KV-cache budget. An attacker able to submit video requests to a video-capable GPU deployment with PyNvVideoCodec installed can exhaust shared GPU memory, causing request failures, worker crashes, or denial of service. The first release containing the fix is version 0.28.0.

Risk

CVSS
6.5
Severity
MEDIUM
Vector
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

Timeline

Published
Sep 16, 2026 18:17
Modified
Oct 7, 2026 14:16
Last observed
Oct 7, 2026 20:05

CWE

CWE-400CWE-770

Affected product identifiers

cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*

References