CVE-2026-69147

vLLM: Request-selected PyNvVideoCodec GPU decode bypasses static VRAM reservation

MEDIUM 6.5

At a glance

Source
NVD
EPSS
EPSS 0.58 % · percentile 48 ()
State
PUBLISHED
Published
Updated
Assigner
GitHub_M
NVD status
Analyzed

Description

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.

Weaknesses (CWE)

  • CWE-400 — Uncontrolled Resource Consumption
  • CWE-770 — Allocation of Resources Without Limits or Throttling

References

Scores

VersionScoreVectorScored by
CVSS 3.1MEDIUM 6.5CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:HGitHub_M
CVSS 3.1MEDIUM 6.5CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:HNVD · security-advisories@github.com (Secondary)

Advisories

European references

Affected configurations (NVD CPE)

Affected products

ProductVersions
vllm-project · vllmaffected < 0.28.0

Sources

SourceLicence
CVE List (CVEProject cvelistV5)CVE Program Terms of Use
EPSS (FIRST.org)FIRST — attribution requested
ENISA EUVDENISA — reproduction authorised, source acknowledged
NVD (NIST) via fkie-cad/nvd-json-data-feedsPublic domain — this product uses data from the NVD API but is not endorsed or certified by the NVD