CVE-2026-69147 vLLM: Request-selected PyNvVideoCodec GPU decode bypasses static VRAM reservation
Attackers can exhaust GPU memory by forcing specific video decoders, causing crashes in vLLM deployments with PyNvVideoCodec installed.
- Exploited in the wild
- Not known
- Login needed
- Any signed-in user
- Attackable
- Over the internet
- Victim action needed
- Not needed
- Fix
- Available
What is CVE-2026-69147?
CVE-2026-69147 is a medium-severity denial of service vulnerability in Vllm Vllm, published on 16 Sep 2026. Any signed-in user can exploit it over the network, with no action from the victim.
Official 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.
Which versions are affected by CVE-2026-69147?
3 open source packages ship an affected version, including vllm. It affects 1 product.
| Package | Affected | Fixed in |
|---|---|---|
vllm PyPI | All versions | 0.28.0 |
vllm PyPI | All versions | 0.28.0 |
vllm PyPI | All versions | 0.28.0 |
Products include Vllm Vllm.
How do I fix CVE-2026-69147?
Upgrade vllm to 0.28.0 or later.
Upgrade, then put limits on request size, rate and nesting depth so one client cannot exhaust the service. More about Denial of service.
Is CVE-2026-69147 being exploited?
It is not confirmed as exploited in the wild, but public exploit code for CVE-2026-69147 exists.
Other vulnerabilities in Vllm Vllm
- CVE-2025-32444 vLLM Vulnerable to Remote Code Execution via Mooncake Integration
- CVE-2026-22778 vLLM leaks a heap address when PIL throws an error
- CVE-2025-47277 vLLM Allows Remote Code Execution via PyNcclPipe Communication Service
- CVE-2024-11041 Remote Code Execution in vllm-project/vllm
- CVE-2026-48746 vLLM: OpenAI auth bypass
Sources
Also known as GHSA-8pw2-6jv3-mj5j, MINI-68cj-x2m2-v375, MINI-6cp8-3j5g-h76r, MINI-rv8r-8878-8rgw, PYSEC-2026-4178.