Vulnerability in Vllm
CVE-2025-62372
vLLM is an inference and serving engine for large language models (LLMs). From version 0.5.5 to before 0.11.1, users can crash the vLLM engine serving multimodal models by passing multimodal embedding inputs with correct ndim but incorrect shape (e.g. hidden dimension is wrong), regardless of whether the model is intended to support such inputs (as defined in the Supported Models page). This issue has been patched in version 0.11.1.
Published · last modified .
CVSS v3 metric
CVSS v3 base score 6.5 (Medium). Vector: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H.
EPSS exploit prediction
EPSS 0.004, 35.8th percentile (scored ); unchanged over 30 days. About EPSS.
Affected products
- Vllm — versions 0.11.1
- Vllm-Project Vllm — versions >= 0.5.5, < 0.11.1
Weakness classification (CWE)
References
- https://github.com/vllm-project/vllm/security/advisories/GHSA-pmqf-x6x8-p7qw (Mitigation, Vendor Advisory)
- https://github.com/vllm-project/vllm/pull/27204 (Patch, Issue Tracking, Vendor Advisory)
- https://github.com/vllm-project/vllm/pull/6613 (Issue Tracking)
- https://github.com/vllm-project/vllm/commit/58fab50d82838d5014f4a14d991fdb9352c9c84b (Patch)