CVE-2026-54232 vLLM: Dependency Confusion Vulnerability in vLLM Dockerfile
Attackers can execute arbitrary code and steal data by registering a missing package on PyPI, exploiting dependency confusion in the vLLM Docker build process.
- Exploited in the wild
- Not known
- Login needed
- Not needed
- Attackable
- Over the internet
- Victim action needed
- A user has to click or open something
- Fix
- Available
What is CVE-2026-54232?
CVE-2026-54232 is a high-severity vulnerability in Vllm Vllm, published on 22 Jun 2026. An attacker can exploit it over the internet without logging in, and only if a user opens a file or follows a link.
Official description
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.
Weakness: CWE-427
Which versions are affected by CVE-2026-54232?
1 open source package ship an affected version, including vllm. It affects 1 product.
| Package | Affected | Fixed in |
|---|---|---|
vllm PyPI | All versions | 0.22.1 |
Products include Vllm Vllm.
How do I fix CVE-2026-54232?
Upgrade vllm to 0.22.1 or later.
Is CVE-2026-54232 being exploited?
There is no known exploitation of CVE-2026-54232 in the wild and no public exploit code.
Other vulnerabilities in Vllm Vllm
- CVE-2025-32444 vLLM Vulnerable to Remote Code Execution via Mooncake Integration
- CVE-2026-22778 vLLM leaks a heap address when PIL throws an error
- CVE-2025-47277 vLLM Allows Remote Code Execution via PyNcclPipe Communication Service
- CVE-2024-11041 Remote Code Execution in vllm-project/vllm
- CVE-2026-48746 vLLM: OpenAI auth bypass
Sources
Also known as GHSA-jrf6-vqxq-pjv2, MINI-cvf6-43v8-9xvp, MINI-wpcq-p45h-gv8h, MINI-x97c-5xgp-qww8, PYSEC-2026-227.