---
id: CVE-2026-69147
title: vLLM is an inference and serving engine for large language models
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 …
severity: medium
cvss: 6.5
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H'
cwe:
  - CWE-400
  - CWE-770
vendor: vllm-project
product: vllm
affected:
  - vllm < 0.28.0
published: '2026-09-16'
updated: '2026-09-16'
sourceUpdated: '2026-09-16T19:17:26.113'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-69147'
references:
  - url: >-
      https://github.com/vllm-project/vllm/commit/283893c72292ede38d277e3cd2b9b64c3e4f1dda
    label: security-advisories@github.com
  - url: >-
      https://github.com/vllm-project/vllm/commit/ba22152096b2484faa3579624a253d54804d876d
    label: security-advisories@github.com
  - url: 'https://github.com/vllm-project/vllm/pull/47259'
    label: security-advisories@github.com
  - url: >-
      https://github.com/vllm-project/vllm/security/advisories/GHSA-8pw2-6jv3-mj5j
    label: security-advisories@github.com
  - url: >-
      https://github.com/vllm-project/vllm/security/advisories/GHSA-8pw2-6jv3-mj5j
    label: 134c704f-9b21-4f2e-91b3-4a467353bcc0
  - url: >-
      https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-69147.json
  - url: 'https://access.redhat.com/security/cve/CVE-2026-69147'
  - url: 'https://bugzilla.redhat.com/show_bug.cgi?id=2535581'
  - url: 'https://www.cve.org/CVERecord?id=CVE-2026-69147'
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-69147'
  - url: 'https://github.com/vllm-project/vllm/releases/tag/v0.25.0'
  - url: 'https://github.com/advisories/GHSA-8pw2-6jv3-mj5j'
  - url: 'https://github.com/vllm-project/vllm'
tags:
  - nvd
  - cve.org
  - exploit-available
  - csaf
  - vex
  - red-hat
  - ghsa
  - pip
  - osv
exploitAvailable: true
ssvc:
  exploitation: poc
  automatable: 'no'
  technicalImpact: partial
  timestamp: '2026-09-16T18:37:08.090589Z'
ingestedAt: '2026-09-16T18:01:17.455Z'
aliases:
  - GHSA-8pw2-6jv3-mj5j
ecosystem: pip
patched:
  - vllm 0.28.0
epss: 0.0046
epssPercentile: 0.39141
---

## Overview

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.

## Remediation

Refer to the linked advisories for vendor-supplied fixes and affected version ranges.

## Vendor advisories

- **Red Hat VEX** · Moderate · affected: Red Hat AI Inference Server, Red Hat Enterprise Linux AI (RHEL AI) 3, Red Hat OpenShift AI (RHOAI) · no fix planned: Red Hat AI Inference Server, Red Hat Enterprise Linux AI (RHEL AI) 3, Red Hat OpenShift AI (RHOAI) · updated 2026-09-16 · [vex](https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-69147.json)

## Package advisory (CVE-2026-69147)

Affected packages:

- `vllm < 0.28.0`

Patched in:

- `vllm 0.28.0`

Source: https://github.com/advisories/GHSA-8pw2-6jv3-mj5j