Home

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.

PUBLISHED Reserved 2026-03-30 | Published 2026-04-02 | Updated 2026-04-03 | Assigner GitHub_M




MEDIUM: 5.9CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:H/A:L

Problem types

CWE-20: Improper Input Validation

Product status

>= 0.5.5, < 0.18.0
affected

References

github.com/...t/vllm/security/advisories/GHSA-6c4r-fmh3-7rh8

github.com/vllm-project/vllm/pull/37058

github.com/...ommit/c7f98b4d0a63b32ed939e2b6dfaa8a626e9b46c4

github.com/vllm-project/vllm/releases/tag/v0.18.0

cve.org (CVE-2026-34760)

nvd.nist.gov (CVE-2026-34760)

Download JSON