Описание
vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.23.1rc0, vLLM's /v1/audio/transcriptions endpoint limits compressed upload size but not decoded PCM output. A 25MB OPUS file expands to ~14.9GB of float32 PCM at decode time. This vulnerability is fixed in 0.23.1rc0.
A flaw was found in vLLM, an inference and serving engine for large language models (LLMs). A remote attacker could exploit a vulnerability in the /v1/audio/transcriptions endpoint. By uploading a specially crafted compressed audio file, such as an OPUS file, the attacker could cause the system to allocate an excessive amount of memory during the decoding process. This uncontrolled memory allocation can lead to a Denial of Service (DoS) condition, making the service unavailable to legitimate users.
Отчет
Red Hat rates this issue as having Moderate impact. A crafted audio upload to the vLLM /v1/audio/transcriptions endpoint can cause excessive decoded PCM allocation and denial of service. Affected components are vLLM serving images in Red Hat AI Inference Server, Red Hat OpenShift AI, and Red Hat Enterprise Linux AI bootc that ship vLLM prior to 0.23.1. KServe sidecars are not affected.
Меры по смягчению последствий
Limit access to audio transcription endpoints. Enforce authentication and request rate limits at the route or gateway. Set pod memory limits and restart policies. Upgrade to vLLM >= 0.23.1 when available from Red Hat.
Затронутые пакеты
| Платформа | Пакет | Состояние | Рекомендация | Релиз |
|---|---|---|---|---|
| Red Hat AI Inference Server | rhaiis/vllm-cpu-rhel9 | Fix deferred | ||
| Red Hat AI Inference Server | rhaiis/vllm-cuda-rhel9 | Fix deferred | ||
| Red Hat AI Inference Server | rhaiis/vllm-neuron-rhel9 | Fix deferred | ||
| Red Hat AI Inference Server | rhaiis/vllm-rocm-rhel9 | Fix deferred | ||
| Red Hat AI Inference Server | rhaiis/vllm-spyre-rhel9 | Fix deferred | ||
| Red Hat AI Inference Server | rhaiis/vllm-tpu-rhel9 | Fix deferred | ||
| Red Hat AI Inference Server | rhaii/vllm-cpu-rhel9 | Fix deferred | ||
| Red Hat AI Inference Server | rhaii/vllm-cuda-rhel9 | Fix deferred | ||
| Red Hat AI Inference Server | rhaii/vllm-gaudi-rhel9 | Fix deferred | ||
| Red Hat AI Inference Server | rhaii/vllm-neuron-rhel9 | Fix deferred |
Показывать по
Дополнительная информация
Статус:
EPSS
6.5 Medium
CVSS3
Связанные уязвимости
vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.23.1rc0, vLLM's /v1/audio/transcriptions endpoint limits compressed upload size but not decoded PCM output. A 25MB OPUS file expands to ~14.9GB of float32 PCM at decode time. This vulnerability is fixed in 0.23.1rc0.
vLLM is an inference and serving engine for large language models (LLM ...
vLLM: OOM Denial of Service via Audio Decompression Bomb
EPSS
6.5 Medium
CVSS3