TI-2026-27004 Aliases CVE-2026-34760 GHSA-6c4r-fmh3-7rh8

vLLM: Downmix Implementation Differences as Attack Vectors Against Audio AI Models

LOW 31CWE-20Published 2026-04-02 · Updated 2026-04-03

Key facts

Attack vector: Network
Privileges required: Low
User interaction: None
Attack complexity: Low
Impact type: —
Confidentiality · integrity · availability: None · High · Low
Fix status: Partial Red Hat: 3 without a fix

Attack profile from CVSS 3.1 · NVD. Fix status from CNA, GitHub and Red Hat.

Summary

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.

Why it matters now

Exploitation None reported CISA SSVC · not in CISA KEV

EPSS 0.00277 20th percentileScored 2026-11-07

Fix Partial vllm (vllm-project): 0.18.0vllm (PyPI): 0.18.0Red Hat: 3 without a fix

SSVC decision points

CISA ADP · SSVC 2.0.3

Exploitation: None
Automatable: No
Technical impact: Partial

The decision (Track, Attend, Act) also depends on your mission and well-being impact; see CISA's SSVC guide.

Affected products

Vendor	Product	Affected versions	Fixed in	Status	Source
vllm-project	vllm	0.5.5 to before 0.18.0	0.18.0	Patch available	CNA
—	vllm (PyPI)	0.5.5 to before 0.18.0	0.18.0	Patch available	GHSA-6c4r-fmh3-7rh8

vllm Vendorvllm-project Affected 0.5.5 to before 0.18.0 Fixed in 0.18.0 Patch available CNA
vllm (PyPI) Vendor— Affected 0.5.5 to before 0.18.0 Fixed in 0.18.0 Patch available GHSA-6c4r-fmh3-7rh8

CPE configurations NVD · 1

CPE	Versions	Role	Vendor
cpe:2.3:a:vllm:vllm::::::::	0.5.5 to before 0.18.0	Vulnerable	vLLM

cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:* Versions 0.5.5 to before 0.18.0 Vulnerable ·vLLM

Remediation

Fix · CNA vllm (vllm-project) Fixed in 0.18.0
Fix · GHSA-6c4r-fmh3-7rh8 vllm (PyPI) Fixed in 0.18.0 Advisory →

Vendor assessments

Red Hat

Red Hat rating: Moderate. Red Hat's page for this CVE →

Product	Status	Plan or fix
Red Hat AI Inference Server	Affected	No fix yet: Fix deferred
Red Hat Enterprise Linux AI (RHEL AI) 3	Affected	No fix yet: Fix deferred
Red Hat OpenShift AI (RHOAI)	Affected	No fix yet: Fix deferred

Timeline

2026-10-02 Threatist record created from CVE-2026-34760
2026-10-03 Alias added: GHSA-6c4r-fmh3-7rh8

References

Patch 1

https://github.com/vllm-project/vllm/commit/c7f98b4d0a63b32ed939e2b6dfaa8a626e9b46c4

Advisory 2

Other 4

TI-2024-10229 Remote Code Execution in vllm-project/vllm CVSS 3.0CRITICAL 9.8
TI-2024-08404 Remote Code Execution in vllm-project/vllm CVSS 3.1CRITICAL 9.8
TI-2026-36017 vLLM: OpenAI auth bypass CVSS 3.1CRITICAL 9.1
TI-2026-63056 vLLM before 0.28.0 Denial of Service via negative token ID CVSS 4.0HIGH 8.7
TI-2026-04308 Hardcoded trust_remote_code=True in vllm-project/vllm Bypasses User Security Control CVSS 3.0HIGH 8.8
TI-2026-39137 vLLM: Remote DoS in vLLM via Invalid Recovered Token Reinjection CVSS 3.1HIGH 7.5

See all 124 →

Sources

This product uses the NVD API but is not endorsed or certified by the NVD. Sources and licences