CVE-2026-34760 in vLLM

摘要 (英语)

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.

Be aware that VulDB is the high quality source for vulnerability data.

负责

GitHub_M

预定

2026-03-30

披露

2026-04-02

状态

已确认

条目

VulDB provides additional information and datapoints for this CVE:

标识符	漏洞	CWE	可利用	对策	CVE
355011	vLLM numpy.mean 权限提升	20	未定义	官方修复	CVE-2026-34760

来源

• MITRE
• NVD
• CVE Details