Goal Reached Thanks to every supporter — we hit 100%!

Goal: 1000 CNY · Raised: 1336 CNY

100%
Buy Us a Coffee

CVE-2026-90554— vLLM before 0.28.0 Denial of Service via audio extraction

Quick assessment

Affected
vllm-project vLLM
Exploitation
No confirmed in-the-wild exploitation; assess based on exposure
Recommended action
Check the vendor advisory and references for a fixed version. If immediate upgrade is impossible, restrict exposure and increase monitoring.

vLLM 版本 0.10.2 及以上但低于 0.28.0 的发行版，在为 NanoNemotronVL 模型从视频输入中提取音频时，未对音频解码大小或时长施加任何限制。 具体而言，在 中， 函数调用 时，未传入 或 参数，因此环境变量 和 所定义的限制并未生效（这与直接上传音频的路径 中的处理方式不同，后者会强制应用这些限制）。 当以 配置部署 NanoNemotronVL 模型时，攻击者可以提供一个体积小但高度压缩的视频作为多模态输入，从而迫使服务器在音频解码阶段分配数 GB 的内存，最终导致服务不可用（DoS

CVSS 6.2 · Medium

Possible ATT&CK Techniques 1 AI

T1499 · Endpoint Denial of Service
Get alerts for future matching vulnerabilities Log in to subscribe

I. Basic Information for CVE-2026-90554

Vulnerability Information
Have questions about the vulnerability? See if Shenlong's analysis helps!
View Shenlong Deep Dive ↗

Although we use advanced large model technology, its output may still contain inaccurate or outdated information.Shenlong tries to ensure data accuracy, but please verify and judge based on the actual situation.

Vulnerability Title
vLLM before 0.28.0 Denial of Service via audio extraction
Source: CVE Program / CVE List V5
Vulnerability Description
vLLM versions >=0.10.2 and <0.28.0 do not apply any audio decode-size or duration limit when extracting audio from video input for NanoNemotronVL models. In nano_nemotron_vl.py, _extract_audio_from_videos calls load_audio_pyav(BytesIO(video_bytes)) without the max_duration_s or max_decode_bytes parameters, so neither VLLM_MAX_AUDIO_DECODE_DURATION_S nor VLLM_MAX_AUDIO_DECODE_BYTES is enforced (unlike the direct audio upload path in AudioMediaIO). When a NanoNemotronVL model is served with use_audio_in_video=True, an attacker who supplies a small, highly compressed video as multimodal input can force the server to allocate gigabytes of memory during audio decoding, resulting in a denial of service. Fixed in vLLM 0.28.0.
Source: CVE Program / CVE List V5
CVSS Information
CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Source: CVE Program / CVE List V5
Vulnerability Type
未加控制的资源消耗(资源穷尽)
Source: CVE Program / CVE List V5

Affected Products

Vendor Product Affected Versions CPE Subscribe
vllm-project vLLM 0.10.2 ~ 0.28.0 -

II. Public POCs for CVE-2026-90554

# POC Description Source Link Shenlong Link
AI-Generated POC Premium

No public POC found.

Login to generate AI POC

III. Intelligence Information for CVE-2026-90554

请登录查看更多情报信息。

Vendor Advisories for CVE-2026-90554 (2)

Same Patch Batch · vllm-project · 2026-09-12 · 3 CVEs total

CVE-2026-90553 7.8 HIGH vLLM before 0.28.0 Remote Code Execution via LlavaOnevision2 processor
CVE-2026-90555 6.5 MEDIUM vLLM before 0.28.0 Denial of Service via Audio Header

IV. Related Vulnerabilities

Same product: vLLM
Same vendor: vllm-project
Same weakness: CWE-400

V. Comments for CVE-2026-90554

No comments yet

Leave a comment