| Component | vLLM |
| Severity | high (CVSS 7.5) |
| CVSS vector | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H |
| Published (NVD) | 2026-10-21 |
| Affected versions | >= 0.24.0, < 0.31.0 read from the advisory text, followed where the structured sources say otherwise |
| Fixed in | 0.31.0 |
| Weakness | CWE-770 |
| CISA KEV | not listed |
NVD description
vLLM after 0.24.0 and before 0.31.0 contains a memory exhaustion vulnerability in MOSS-Audio model serving that caches a processor for every distinct request-controlled mm_processor_kwargs value in an unbounded dictionary. Unauthenticated attackers can send distinct or oversized mel_config values to the /tokenize endpoint, which bypasses --api-key, to exhaust host memory and crash the API process.
What to do
- Upgrade vLLM to
0.31.0or later. - Do not expose the component's API or UI to the internet without authentication in front of it.
Sources
- NVD: CVE-2026-114052
- https://github.com/vllm-project/vllm/security/advisories/GHSA-823j-m4cj-hmjf
- https://www.vulncheck.com/advisories/vllm-0.25.0-before-0.31.0-moss-audio-memory-exhaustion-via-tokenize-mm-processor-kwargs