vLLM: Unauthenticated audio decompression-bomb DoS in /v1/chat/completions
Summary
The audio decode-duration guard (max_duration_s, env VLLM_MAX_AUDIO_DECODE_DURATION_S, default 600s) that protects against audio decompression-bomb DoS is wired into only the speech-to-text path (/v1/audio/transcriptions). The chat audio path (/v1/chat/completions, input_audio content parts) calls the same decoder with no limit, so an unauthenticated client can submit a few-KB compressed audio file that expands to multiple GB of float32 PCM at decode time, OOM-killing the worker. This is a distinct sibling of CVE-2026-5497 (video frame-count bomb, VideoMediaIO.load_base64) and GHSA-pq5c-rjhq-qp7p (image) in the same media subsystem.
Verified against main at HEAD d78650c (2026-06-16); applicable to the latest release v0.23.0.
Details
The guard rejects long audio during decode (before allocation), implemented in vllm/multimodal/media/audio.py:
load_audio_pyav— metadata reject (~82-98) and live sample-count reject (~129-136)load_audio_soundfile— frames reject (~165-174)
All are gated on if max_duration_s is not None.
It is passed in exactly one place — the transcription serving layer:
# .../speech_to_text/base/serving.py:~170-174
load_audio(buf, sr=..., max_duration_s=self.max_audio_decode_duration_s)
# self.max_audio_decode_duration_s = envs.VLLM_MAX_AUDIO_DECODE_DURATION_S (default 600)
The chat path never threads it:
# vllm/multimodal/media/audio.py:237-238
def load_bytes(self, data: bytes) -> tuple[npt.NDArray, float]:
return load_audio(BytesIO(data), sr=None) # no max_duration_s -> every guard above is skipped
Unauthenticated reachability chain (chat):
parse_input_audio (chat_utils.py) -> parse_audio -> connector.fetch_audio -> AudioMediaIO._load_data_url -> load_base64 -> load_bytes -> load_audio(..., sr=None). The connector never passes max_duration_s, and inline data: URLs need no HTTP fetch (so VLLM_AUDIO_FETCH_TIMEOUT does not bound them). The OpenAI-compatible server has no auth by default (auth only when --api-key / VLLM_API_KEY is set).
Impact
Unauthenticated remote denial of service (availability) via memory amplification on a default-no-auth endpoint, on any deployment serving an audio-capable model. Same class and impact as the sibling CVE-2026-5497 (video). CWE-770 / CWE-409.
Fix
A fix was introduced in this MR: https://github.com/vllm-project/vllm/pull/45908
Czego dotyczy
Kliknij technologię, aby zobaczyć wszystkie dotyczące jej podatności.
| Platforma | Paczka | Podatne wersje | Poprawka w wersji |
|---|---|---|---|
| Python (PyPI) | vllm | < 0.24.0 | 0.24.0 |
Odnośniki
- https://github.com/vllm-project/vllm/security/advisories/GHSA-hcwq-8wjf-3gcr
- https://nvd.nist.gov/vuln/detail/CVE-2026-57173
- https://github.com/vllm-project/vllm/pull/45908
- https://github.com/vllm-project/vllm/commit/3d20275bb4d434f53055c3c0b645fd8bb072965e
- https://github.com/vllm-project/vllm
- https://github.com/vllm-project/vllm/releases/tag/v0.24.0
- https://pypi.org/project/vllm
- https://github.com/advisories/GHSA-hcwq-8wjf-3gcr
- Wektor CVSS
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H- Słabość (CWE)
CWE-770- Platforma
- Python (PyPI)
- Paczka
vllm- Ujawniono
- 16.09.2026
- Aktualizacja
- 11.10.2026 15:18
Źródła i licencje
OSV / GitHub Advisory Database - PYSEC-2026-4179 - PyPA Advisory Database; CC BY 4.0
OSV / GitHub Advisory Database - GHSA-hcwq-8wjf-3gcr - CC BY 4.0
Historia podatności: vllm
Wszystkie opisane podatności tej paczki (Python (PyPI)), od najnowszej. Widać, czy to jednorazowa wpadka, czy paczka, której trzeba pilnować.
- Wszystkich wpisów
- 94
- Pierwsza → ostatnia
- 09.2024 10.2026
- Bez poprawki
- 5
- Wykorzystywana w atakach
- 0
Rozkład istotności
- Krytyczna 9
- Wysoka 23
- Średnia 47
- Niska 4
- Nieoceniona 11
Ujawnienia w latach
- 2024: 2
- 2025: 30
- 2026: 62
-
vLLM: Unauthenticated audio decompression-bomb DoS in /v1/chat/completions ten wpis
Dowiedz się, które podatności dotyczą Twojej aplikacji
Po podłączeniu integracji DockRay i włączeniu raportowania oprogramowania DockRay porównuje zgłoszone wersje komponentów ze znanymi podatnościami i powiadamia o potwierdzonych dopasowaniach. Zakres zależy od obsługiwanej technologii i dostępnych biuletynów.