CVE-2026-34760
Low
No strong exploitation signal.
CVSS base
5.9
MEDIUM
CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:H/A:L
EPSS — probability of exploitation (30 days)
0.5%
38.5th percentile
CISA KEV
Not listed
Weakness / dates
CWE-20
Published 2026-04-02 · modified 2026-07-24
CVSS breakdown
CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:H/A:L
| Attack Vector | N | Network |
| Attack Complexity | H | High |
| Privileges Required | L | Low |
| User Interaction | N | None |
| Scope | U | Unchanged |
| Confidentiality | N | None |
| Integrity | H | High |
| Availability | L | Low |
Timeline
- 2026-04-02 — Published (NVD)
- 2026-07-24 — Last modified (NVD)
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.