mimo-vision-mcp
A lightweight MCP stdio server that enables LLM agents to invoke the MiMo Vision model for local image understanding and OCR via the OpenCode API.
README
MiMo Vision MCP Server
A lightweight Model Context Protocol (MCP) stdio server that enables LLM agents (such as Codex CLI/Desktop, Claude Desktop) to invoke the MiMo Vision model (mimo-v2.5) for local image understanding and OCR.
架构流程 (Architecture)
Codex / Claude (Client)
│ (JSON-RPC 2.0 via stdio)
▼
src/mimo_vision_mcp.py
│ (curl.exe with stream base64 payload)
▼
https://opencode.ai/zen/go/v1/chat/completions (model: mimo-v2.5)
│ (JSON Response)
▼
Recognition / Description Text returned to Client
Why
curl.exe?
Some API gateways reject standard Python HTTP user agents / connection handshakes. Using native systemcurl.exeensures robust connections.
目录结构 (Project Structure)
mimo-vision-mcp/
├── src/
│ └── mimo_vision_mcp.py # MCP stdio 服务端核心实现(纯标准库,零三方依赖)
├── tests/
│ └── test_mcp.py # 自动化测试脚本(生成测试图并验证协议连通性)
├── .env.example # 环境变量配置模版
├── .gitignore # Git 忽略配置
├── pyproject.toml # Python 打包与元数据配置
├── LICENSE # MIT 开源协议
└── README.md # 项目文档
环境变量配置 (Environment Variables)
| 变量名 | 默认值 | 说明 |
|---|---|---|
MIMO_API_KEY |
(必填) | MiMo / OpenCode API 密钥 |
MIMO_API_BASE |
https://opencode.ai/zen/go/v1/chat/completions |
Chat Completions 接口地址 |
MIMO_MODEL |
mimo-v2.5 |
视觉模型名称 |
MIMO_DEFAULT_PROMPT |
请详细描述这张图片的内容。 |
默认识别提示词 |
MIMO_CURL |
curl.exe |
用于发起 HTTP 请求的 curl 可执行程序 |
工具接口 (Tool Definition)
mimo_vision
向 MiMo 视觉模型发送本地图片,返回文字识别或描述结果。
- 参数:
image_path(string, 必填): 本地图片文件的绝对路径(支持.png,.jpg,.jpeg,.webp,.gif,.bmp等)。prompt(string, 可选): 自定义提示词,如"识别图中的文字"、"这张图里有什么物体?"。model(string, 可选): 模型名称覆盖,默认mimo-v2.5。
安装与接入配置 (Setup Guide)
1. Codex 接入配置 (~/.codex/config.toml)
在你的 config.toml 中添加以下配置:
[mcp_servers.mimo_vision]
type = "stdio"
command = "python"
args = ["C:\\Users\\user\\Documents\\Codex\\2026-08-17\\mimo-vision-mcp\\src\\mimo_vision_mcp.py"]
startup_timeout_sec = 120
[mcp_servers.mimo_vision.env]
MIMO_API_KEY = "your-api-key"
MIMO_API_BASE = "https://opencode.ai/zen/go/v1/chat/completions"
MIMO_MODEL = "mimo-v2.5"
2. Claude Desktop 接入配置 (claude_desktop_config.json)
{
"mcpServers": {
"mimo-vision": {
"command": "python",
"args": ["C:/Users/user/Documents/Codex/2026-08-17/mimo-vision-mcp/src/mimo_vision_mcp.py"],
"env": {
"MIMO_API_KEY": "your-api-key",
"MIMO_API_BASE": "https://opencode.ai/zen/go/v1/chat/completions",
"MIMO_MODEL": "mimo-v2.5"
}
}
}
}
运行测试 (Testing)
$env:MIMO_API_KEY="your-api-key"
python tests/test_mcp.py
许可协议 (License)
本项目采用 MIT License 授权。
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.