Disclosure summary
vLLM is an inference and serving engine for large language models. Prior to 0.24.0, the input_audio handling path for /v1/chat/completions calls AudioMediaIO.load_bytes or AudioMediaIO.load_file without passing VLLM_MAX_AUDIO_DECODE_DURATION_S to the shared audio decoder. An unauthenticated client can therefore submit a small compressed audio input that expands into a very large float32 PCM allocation, bypassing the duration guard already used by /v1/audio/transcriptions and causing an out-of-memory worker crash. Inline data URLs reach this path without being bounded by VLLM_AUDIO_FETCH_TIMEOUT. The issue affects deployments serving an audio-capable model, and authentication changes only the deployment-specific reachability. This issue is fixed in version 0.24.0.
Source-reported weakness categories
CWE-770
Source-specific records & product guidance
Sources retain their own attribution and scoring. Follow the original record to confirm affected versions, fixed releases, and configuration conditions.
NIST National Vulnerability Database · NVD-CVE-2026-57173
Open original source · Updated Oct 07, 2026
Only CPE matches marked vulnerable=true are indexed. AND/OR platform conditions must be checked in the original NVD record.
| Vendor | Product / association | Version / bounds |
|---|---|---|
| vllm | vllm | * {"versionEndExcluding":"0.24.0"} |
Original records & references
- NIST NVD record
- CVE Program record
- github.com — Patch
- github.com — Issue Tracking, Patch
- github.com — Release Notes
- github.com — Vendor Advisory
PUBLISHED 2026-09-16T13:17:24-04:00
MODIFIED 2026-10-07T10:24:22-04:00
INGESTED 2026-10-10T20:25:12-04:00