# vLLM

vLLM is a high-performance, easy-to-use engine for LLM inference and serving. For an in-depth overview, see the [official documentation](https://docs.vllm.ai/).

This directory contains vLLM deployment guides for different MiniCPM versions. Use the links below to jump to the specific version.

## Versioned Deployment Guides

| Version | English | 中文 |
| :--- | :---: | :---: |
| **MiniCPM-V 4.6** *(latest)* | [Guide](./minicpm-v4_6_vllm.md) | [指南](./minicpm-v4_6_vllm_zh.md) |
| MiniCPM-V 4.5 | [Guide](./minicpm-v4_5_vllm.md) | [指南](./minicpm-v4_5_vllm_zh.md) |
| MiniCPM-V 4.0 | [Guide](./minicpm-v4_vllm.md)   | [指南](./minicpm-v4_vllm_zh.md)   |
| MiniCPM-o 4.5 | [Guide](./minicpm-o4_5_vllm.md) | [指南](./minicpm-o4_5_vllm_zh.md) |
| MiniCPM-o 2.6 | [Guide](./minicpm-o2_6_vllm.md) | [指南](./minicpm-o2_6_vllm_zh.md) |
| MiniCPM-V 2.6 | [Guide](./minicpm-v2_6_vllm.md) | [指南](./minicpm-v2_6_vllm_zh.md) |
| MiniCPM-V 2.5 | [Guide](./minicpm-v2_5_vllm.md) | [指南](./minicpm-v2_5_vllm_zh.md) |