Medium · CVSS 6.9CVE-2026-90554

vLLM before 0.28.0 Denial of Service via audio extractionCVE-2026-90554

Answer in brief

CVE-2026-90554 records a Medium severity (CVSS 6.9) vulnerability in vLLM before 0.28.0 Denial of Service via audio extraction. The current sources do not mark it as known exploited. The current feed maps vllm-project/vLLM (generic). Check affected ranges and fixed versions before updating.

Analysis pending evidence review

HOL Guard separates source facts from reviewed analysis. See the methodology.

Published Sep 12, 2026Updated Sep 24, 2026Source checked Oct 27, 2026First seen by HOL Sep 12, 2026Material review Sep 24, 2026

Upstream Advisory

Key facts

Risk: Medium · CVSS 6.9
Exploitation: Not marked as known exploited
Affected software: 2 mapped packages or products
Fix availability: Available

Why this deserves its current priority

CVSS is 6.9. The current sources do not mark it as known exploited. Treat this as a source-backed prioritization signal, not a statement about your environment.

Analysis status

Analysis pending evidence review

Factual feed record only; HOL analysis is not approved for indexing. Read the methodology.

Affected scope and exposure questions

The current feed maps vllm-project/vLLM (generic). Check affected ranges and fixed versions before updating.

Affected products and reported versions
Product	Affected versions	Fixed versions
cpe:2.3:a:vllm:vllm::::::::	Not reported	Not reported

Mapped affected packages and fixed versions
Package	Affected range	Fixed version
vllm-project/vLLMgeneric	>=0.10.2 <0.28.0	0.28.0

Recommended response

Check inventory. Check lockfiles and deployed manifests for vllm-project/vLLM.
Review the reported fix. Update vllm-project/vLLM to 0.28.0 if you use the affected versions. Test the change in a non-production environment first.

Evidence timeline and material changes

Published upstream
Sep 12, 2026
Evidence: source:cvelist:source_dates:source-dates:record
Source modified
Sep 24, 2026
Evidence: source:cvelist:source_dates:source-dates:record
First seen by HOL
Sep 12, 2026

Sources and claim methodology

Upstream source 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.

Quoted source text, attributed separately from HOL analysis.