Description
vLLM is an inference and serving engine for large language models (LLMs). From version 0.5.5 to before version 0.18.0, Librosa defaults to using numpy.mean for mono downmixing (to_mono), while the international standard ITU-R BS.775-4 specifies a weighted downmixing algorithm. This discrepancy results in inconsistency between audio heard by humans (e.g., through headphones/regular speakers) and audio processed by AI models (Which infra via Librosa, such as vllm, transformer). This issue has been patched in version 0.18.0.
Affected products
vllm: vllm
References4
- github.com/vllm-project/vllm/security/advisories/GHSA-6c4r-fmh3… Vendor Advisory
- github.com/vllm-project/vllm/commit/c7f98b4d0a63b32ed939e2b6dfa… Patch
- github.com/vllm-project/vllm/pull/37058 Issue Tracking
- github.com/vllm-project/vllm/releases/tag/v0.18.0 Release Notes
Metrics
- CVSS 3.1
-
7.1
HIGH
Score source: NVD
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:L - EPSS — exploit probability
- 0% percentile: 20 · as of 12 November 2026
- CWE
- CWE-20
Could this be exploited in your environment?
A patch closes the known hole. A penetration test shows what an attacker can still reach — with proof.