Cyber Resilience

CVE-2026-54232

HighPublic PoC

Published: 22 June 2026

Published
22 June 2026
Modified
24 June 2026
KEV Added
Patch
CVSS Score v3.1 8.8 CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
EPSS Score 0.0030 22.1th percentile
Risk Priority 55 floored blend · peak EPSS

Summary

CVE-2026-54232 is a high-severity Uncontrolled Search Path Element (CWE-427) vulnerability in Vllm Vllm. Its CVSS base score is 8.8 (High).

Operationally, ranked at the 22.1th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog; a public proof-of-concept is referenced.

OWASP Top 10 for Web (2025)

EU & UK References

Vulnerability details

vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.1, the vLLM Dockerfile is vulnerable to a dependency confusion attack through the flashinfer-jit-cache package. The package is installed from a custom index (flashinfer.ai/whl/) using --extra-index-url,…

more

but the package name was not registered on PyPI, and UV_INDEX_STRATEGY="unsafe-best-match" is set globally. An attacker who registers flashinfer-jit-cache on PyPI with version 0.6.11.post2 can execute arbitrary code as root during the Docker build and backdoor every resulting container image, enabling exfiltration of all user prompts, API credentials, and model data from production vLLM deployments This vulnerability is fixed in 0.22.1.

CWE(s)

Related Threats

MITRE ATT&CK Enterprise TechniquesAI

Insufficient information to map techniques.
Confidence: LOW · MITRE ATT&CK Enterprise v19.0

Affected Assets

vllm
vllm
≤ 0.22.1

Mitigating Controls

No mitigating controls mapped yet. The per-CVE control annotator has not reached this CVE.

References