CVE-2026-69147
- Source
- https://cve.org/CVERecord?id=CVE-2026-69147
- Import Source
- https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2026-69147.json
- JSON Data
- https://api.osv.dev/v1/vulns/CVE-2026-69147
- Aliases
- Downstream
- Published
- 2026-09-16T17:49:20Z
- Modified
- 2026-09-18T03:48:28Z
- Severity
-
- 6.5 (Medium) CVSS_V3 - CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H CVSS Calculator
- Summary
- vLLM: Request-selected PyNvVideoCodec GPU decode bypasses static VRAM reservation
- Details
-
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.
- Database specific
{ "cna_assigner": "GitHub_M", "cwe_ids": [ "CWE-400", "CWE-770" ], "osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/69xxx/CVE-2026-69147.json" }- References
-
- https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/69xxx/CVE-2026-69147.json
- https://github.com/vllm-project/vllm/security/advisories/GHSA-8pw2-6jv3-mj5j
- https://nvd.nist.gov/vuln/detail/CVE-2026-69147
- https://github.com/vllm-project/vllm/commit/283893c72292ede38d277e3cd2b9b64c3e4f1dda
- https://github.com/vllm-project/vllm/commit/ba22152096b2484faa3579624a253d54804d876d
- https://github.com/vllm-project/vllm/pull/47259