Medium severity5.9NVD Advisory· Published Apr 2, 2026· Updated Jul 24, 2026
CVE-2026-34760
CVE-2026-34760
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.
AI Insight
LLM-synthesized narrative grounded in this CVE's description and references.
Affected packages
Versions sourced from the GitHub Security Advisory.
| Package | Affected versions | Patched versions |
|---|---|---|
vllmPyPI | >= 0.5.5, < 0.18.0 | 0.18.0 |
Affected products
3Patches
Vulnerability mechanics
References
7- github.com/vllm-project/vllm/commit/c7f98b4d0a63b32ed939e2b6dfaa8a626e9b46c4nvdPatchWEB
- github.com/advisories/GHSA-6c4r-fmh3-7rh8ghsaADVISORY
- github.com/vllm-project/vllm/security/advisories/GHSA-6c4r-fmh3-7rh8nvdVendor AdvisoryWEB
- nvd.nist.gov/vuln/detail/CVE-2026-34760ghsaADVISORY
- github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2026-2299.yamlghsaWEB
- github.com/vllm-project/vllm/pull/37058nvdIssue TrackingWEB
- github.com/vllm-project/vllm/releases/tag/v0.18.0nvdRelease NotesWEB
News mentions
0No linked articles in our index yet.