CVE-2026-100650

vLLM through 0.29.0 fetches and fully materializes remote or inline media before enforcing its documented media controls (the VLLM_MAX_AUDIO_CLIP_FILESIZE_MB compressed-audio size cap, default 25 MB, and the per-modality --limit-mm-per-prompt item limits). Across four ingress paths — the shared media-acquisition layer (HTTPConnection.get_bytes()/async_get_bytes()), the chat completions audio_url/base64 path, the batch speech runner, and the Rust frontend POST /tokenize route — the server reads the entire HTTP response body, base64-decodes the inline payload, or spawns one fetch/decode task per media part, and only then applies the limit (or, on some paths, never applies it). A remote attacker can therefore cause the API server or batch-runner process to allocate memory and consume outbound bandwidth proportional to an attacker-chosen body size or media item count before the request is rejected, resulting in pre-inference memory and bandwidth exhaustion (denial of service). The chat and batch surfaces require an API key when one is configured; the Rust frontend /tokenize route is unauthenticated by design. There is no code execution or data disclosure impact.

CVSS 4.0

7.1CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N

CVSS 3.1

6.5CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

EPSS

0.8% chance of exploitation in the next 30 days; higher than 56% of CVEs (scores of 2026-10-25)

CWE

CWE-400

Published

2026-09-26

Modified

2026-10-06

Record from

the NVD, synced 2026-10-19

Affected

• vllm:vllm

Packages (OSV)

References

• https://github.com/vllm-project/vllm/commit/752a3a504485790a2e8491cacbb35c137339ad34
• https://github.com/vllm-project/vllm/security/advisories/GHSA-p6g9-7v3x-m8mv
• https://www.vulncheck.com/advisories/vllm-before-0.29.0-resource-exhaustion-via-unbounded-media-materialization
• https://github.com/vllm-project/vllm/security/advisories/GHSA-p6g9-7v3x-m8mv

Check your stackJSON