CVE-2026-57173: vLLM input_audio path allows memory exhaustion

Summary

vLLM versions prior to 0.24.0 have an input_audio handling flaw on the /v1/chat/completions endpoint. The path calls AudioMediaIO.load_bytes or AudioMediaIO.load_file without passing VLLM_MAX_AUDIO_DECODE_DURATION_S to the shared audio decoder, so the duration guard used by /v1/audio/transcriptions is bypassed. An unauthenticated client can submit a small compressed audio input that expands into a very large float32 PCM allocation, causing an out-of-memory worker crash. Inline data URLs are also not bounded by VLLM_AUDIO_FETCH_TIMEOUT. The issue affects deployments serving an audio-capable model, and authentication changes only deployment-specific reachability.

Mitigation

Fixed in version 0.24.0.

Affected packages in the Exposure Registry

Matched by package name and ecosystem. Each entry shows whether the package delegates to a language model and how many tracked packages depend on it.

• vllmPyPILLM dependency since 2023-08-25 · 15 tracked dependents

Topics

• Inference infrastructure

Related items

• LowCloakFEM: Agentic FEM Code Generation for Red-Teaming Computational Resource ExhaustionSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• MediumThe Rise of LLMJacking: How Threat Actors Monetize Compromised Cloud and Model ResourcesSimilar attack · Google Threat Intelligence
• InfoThe Blind Spots of Detection: Adversarial Perturbation Attacks against Watchdogs in Distributed LLM InferenceSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• HighCVE-2026-118761: Kubeflow Pipelines decompression bomb in pipeline archive uploadSimilar attack · NVD/CVE Database
• LowOpenCode User Hit with Hostile Backlash After Reporting Free AI LoopholeSimilar attack · Socket Blog