CVE-2025-62372
MEDIUMDescription
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.
Is your site exposed to CVE-2025-62372?
Run a free security scan — no signup, results in seconds.
CVSS v3.1 Score
Weakness Type (CWE)
Affected Products
| Vendor | Product |
|---|---|
| vllm | vllm |
| vllm | vllm |
| vllm | vllm |
References
Frequently Asked Questions
What is CVE-2025-62372? +
How severe is CVE-2025-62372? +
What products are affected by CVE-2025-62372? +
How do I check if I'm vulnerable to CVE-2025-62372? +
Related Vulnerabilities
A weakness in Automated Logic and Carrier i-Vu Gen5 router on driver version drv_gen5_106-01-2380, allows malformed packets to be sent …
Taiko Alethia is an Ethereum-equivalent, permissionless, based rollup designed to scale Ethereum without compromising its fundamental properties. In 2.3.1 and …
Improper validation in Power Management Firmware (PMFW) may allow an attacker with privileges to pass malformed workload arguments when exporting …
Memory corruption while parsing the ML IE due to invalid frame content.
Vyper is a Pythonic Smart Contract Language for the Ethereum Virtual Machine. Arrays can be keyed by a signed integer, …
FFmpeg version n6.1 was discovered to contain an improper validation of array index vulnerability in libavcodec/cbs_h266_syntax_template.c. This vulnerability allows attackers …