CVE-2026-54232
1 advisory.
PyPI/vllm CVE-2026-54232: high 8.8, EPSS 0.56%
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.
Action: Upgrade vllm to 0.22.1 or later.
pip install vllm==0.22.1Affected
| introduced | fixed | last affected |
|---|---|---|
| 0 | 0.22.1 | - |
Fixed versions
| 0.22.1 |
Sources
CVE-2026-54232 · published · modified · JSON