CVE-2025-62372
vLLM vulnerable to DoS with incorrect shape of multimodal embedding inputs
- State
- Published
- Severity
- 8.3 High
- Published
- Nov 21, 2025
- Updated
- Nov 24, 2025
- Reserved
- Oct 10, 2025
Description
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.
Scores
| CVSS | Score | Vector | Source |
|---|---|---|---|
| 4.0 | 8.3 High | CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:H | GitHub_M |
Weaknesses
- CWE-129 Improper Validation of Array Index
Affected Products
| Product | Versions |
|---|---|
vllmvllm-project |
|
References
Every record here comes from the CVE Program’s CVE List. © 1999–2026 The MITRE Corporation. CVE is a registered trademark of The MITRE Corporation. The records are reproduced under the CVE Program terms of use.