Improper input validation in Vllm
CVE-2026-34760
- CVSS v3
- 5.9 Medium
- Public exploits
- None known
- EPSS
- 0.003 · 21.4th percentile
- Published
- 2026-04-02
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 · last modified .
CVSS v3 metric
CVSS v3 base score 5.9 (Medium). Vector: CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:H/A:L.
- Attack Vector
- Network
AV:N - Attack Complexity
- High
AC:H - Privileges Required
- Low
PR:L - User Interaction
- None
UI:N - Scope
- Unchanged
S:U - Confidentiality
- None
C:N - Integrity
- High
I:H - Availability
- Low
A:L
EPSS exploit prediction
EPSS 0.003, 21.4th percentile (last changed ); unchanged over 30 days. About EPSS.
Affected products
- Vllm
- Vllm-Project Vllm — versions >= 0.5.5, < 0.18.0
Weakness classification (CWE)
References
- https://github.com/vllm-project/vllm/security/advisories/GHSA-6c4r-fmh3-7rh8 (Vendor Advisory)
- https://github.com/vllm-project/vllm/pull/37058 (Issue Tracking)
- https://github.com/vllm-project/vllm/commit/c7f98b4d0a63b32ed939e2b6dfaa8a626e9b46c4 (Patch)
- https://github.com/vllm-project/vllm/releases/tag/v0.18.0 (Release Notes)