CVE-2026-54232
highpriority · track · 0EPSS 0.006 · 47 %ileDEEUUSEPSS · scored 2026-10-27
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.
cvss
Technical severity, 0–10. One row per scoring source.
base score 0–10 · higher = more severe · hover a vector to decode
ssvc
A stakeholder decision — act, attend, or track — from exploitation and impact.
weaknesses
The CWE weakness classes this flaw belongs to.
affected products (5)
Products and version ranges known to be affected.
advisories (2)
Vendor and ecosystem advisories describing this.
- WID-SEC-W-2026-1860vllm: Schwachstelle ermöglicht Ausführen von beliebigem Programmcode mit AdministratorrechtenDeutsch2026-06-09csaf · cert_bundDE
Ein entfernter, anonymer Angreifer kann eine Schwachstelle in vllm ausnutzen, um beliebigen Programmcode mit Administratorrechten auszuführen.
- PYSEC-2026-2272026-06-22vendor · osvUS
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.
also known as (4)
Other names and identifiers for this entity.
sources
Which sources assert this record, and when.
references (4)
External citations — patches, advisories, analysis.
- http://github.com/vllm-project/vllm/releasesPatchcnnvd
- https://github.com/vllm-project/vllm/security/advisories/GHSA-jrf6-vqxq-pjv2cnnvd
- https://github.com/vllm-project/vllm/security/advisories/GHSA-jrf6-vqxq-pjv2enisa_euvd
- https://github.com/vllm-project/vllm/security/advisories/GHSA-jrf6-vqxq-pjv2ExploitThird Party Advisorynvd
Attackers weaponise vulnerabilities like this against internet-exposed services. See what is exposed on your perimeter with a free passive scan.
Free scan