CVE-2025-62372
Last modified
CVE-2025-62372 is a medium-severity vulnerability rated 6.5/10 on the CVSS scale. 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. EPSS estimates a 0.33% chance of exploitation in the next 30 days.
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.
Metrics
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
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/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
Weakness Enumeration
Affected Software
| Vendor | Product | Versions | Update |
|---|---|---|---|
| Vllm | Vllm | >= 0.5.5, < 0.11.1 | — |
| Vllm | Vllm | 0.11.1 | Rc0 |
References
- https://github.com/vllm-project/vllm/pull/27204Issue Tracking, Patch, Vendor Advisory
- https://github.com/vllm-project/vllm/pull/6613Issue Tracking
- https://github.com/vllm-project/vllm/security/advisories/GHSA-pmqf-x6x8-p7qwMitigation, Vendor Advisory
Timeline
- Published
- Last Modified
- Status
- Analyzed
Frequently Asked Questions
What is CVE-2025-62372?
How severe is CVE-2025-62372?
How do I fix CVE-2025-62372?
How Strix Helps
- How Strix found a critical auth bypass in etcdStrix autonomously discovered a critical authentication bypass in etcd, later designated CVE-2026-33413.
- Autonomous PentestingAI agents that find and validate exploitable vulnerabilities like this one across your applications.
- PR ReviewsPentest every pull request so vulnerable code is caught before it ships to production.
- AI Penetration TestingHow AI-driven penetration testing continuously covers your attack surface.
Related CVEs from 2025
- CVE-2025-62367Taiga is an open source project management platform. In vers…4.8
- CVE-2025-62368Taiga is an open source project management platform. In vers…9
- CVE-2025-62369Xibo is an open source digital signage platform with a web c…7.2
- CVE-2025-6237A vulnerability in invokeai version v6.0.0a1 and below allow…9.8
- CVE-2025-62370Alloy Core libraries at the root of the Rust Ethereum ecosys…7.5
- CVE-2025-62371OpenSearch Data Prepper as an open source data collector for…7.4
- CVE-2025-62373Pipecat is an open-source Python framework for building real…9.8
- CVE-2025-62374Parse Javascript SDK provides access to the powerful Parse S…6.4
- CVE-2025-62375go-witness and witness are Go modules for generating attesta…6.9
- CVE-2025-62376pwn.college DOJO is an education platform for learning cyber…9.5
- CVE-2025-62378CommandKit is the discord.js meta-framework for building Dis…6.1
- CVE-2025-62379Reflex is a library to build full-stack web apps in pure Pyt…3.1
Are you affected by CVE-2025-62372?
Run a free Strix scan to check your systems for this vulnerability.
Scan your code nowSource: NVD / NIST