Advisory

PYSEC-2026-227

Also known as: CVE-2026-54232, GHSA-jrf6-vqxq-pjv2

Summary

The record has no summary.

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, 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.

Severity

High Basis: CVSS v3 score 8.8, vector CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Dates

Published
2026-06-22 23:16 UTC
Modified
2026-06-26 00:00 UTC

Affected packages and fixes

vllm

PyPI · All systems

Affected

  • introduced 0, fixed 0.22.1 (ecosystem)

Fixed versions

0.22.1

Fix

  • Fixed version 0.22.1
    pip install --upgrade vllm==0.22.1

Fix commands are the export's own generic commands. Ojuju never runs them. Check them against your setup first.

Check your own computer: guard packages check

View this record on OSV.dev

Source

Source: PyPA Advisory Database · CC BY 4.0

Sources & licences