1 line
5.4 KiB
JSON
1 line
5.4 KiB
JSON
{"repo": "chanthmiao/mimo-multimodal-understanding-mcp", "files": {"README.md": {"7d4b815654fd8fe077e941ff71611e4f34bdf298": {"t": 1785595221.8875053, "n": 5086, "h": "a9a3f1e960c497b6", "c": "# MiMo Multimodal Understanding MCP Server\n\nMCP server for Xiaomi MiMo multimodal understanding API (image, audio, video).\n\n## Features\n\n- **Image Understanding**: Single/multiple images, URL and local file support\n- **Audio Understanding**: Single/multiple audio, URL and local file support\n- **Video Understanding**: Single/multiple video, URL and local file support, configurable fps and resolution\n\n## Setup\n\n### 1. Install dependencies\n\n```bash\nuv sync\n```\n\n### 2. Configure API Key\n\nCopy `.env.example` to `.env` and fill in your API key:\n\n```bash\ncp .env.example .env\n```\n\nOr set environment variable directly:\n\n```bash\nexport MIMO_API_KEY=your_api_key_here\n```\n\nGet your API key from: https://platform.xiaomimimo.com\n\n### 3. (Optional) Configure API Base URL\n\nThe default API endpoint is determined by your API key prefix:\n\n| Key Prefix | Default Endpoint |\n|------------|------------------|\n| `tp-*` | `https://token-plan-cn.xiaomimimo.com/v1` |\n| `sk-*` or others | `https://api.xiaomimimo.com/v1` |\n\nTo use a different API endpoint:\n\n```bash\nexport MIMO_API_BASE=https://your-custom-endpoint/v1\n```\n\nOr add it to your `.env` file:\n\n```\nMIMO_API_BASE=https://your-custom-endpoint/v1\n```\n\n## Usage\n\n### Quick Start (with uvx)\n\n```bash\nexport MIMO_API_KEY=your_api_key_here\nuvx mimo-multimodal-mcp\n```\n\n### Development mode (with MCP Inspector)\n\n```bash\nuv run mcp dev src/mimo_multimodal_mcp/server.py\n```\n\n### Install to Claude Desktop\n\n```bash\nuv run mcp install src/mimo_multimodal_mcp/server.py\n```\n\n### Direct execution\n\n```bash\nuv run python src/mimo_multimodal_mcp/server.py\n```\n\n### Claude Desktop Configuration\n\nAdd to `~/.config/claude/claude_desktop_config.json`:\n\n```json\n{\n \"mcpServers\": {\n \"mimo-multimodal\": {\n \"command\": \"uvx\",\n \"args\": [\"mimo-multimodal-mcp\"],\n \"env\": {\n \"MIMO_API_KEY\": \"your_api_key_here\"\n }\n }\n }\n}\n```\n\n## Tools\n\n### `understand_image`\n\nAnalyze images using Xiaomi MiMo multimodal model.\n\n| Parameter | Type | Required | Description |\n|-----------|------|----------|-------------|\n| `prompt` | string | Yes | Image understanding task description |\n| `image_url` | string | No | Single image URL or data:image base64 |\n| `image_path` | string | No | Single local image file path |\n| `image_urls` | list[string] | No | Multiple image URLs |\n| `image_paths` | list[string] | No | Multiple local image file paths |\n| `system_prompt` | string | No | Custom system prompt |\n| `max_tokens` | integer | No | Max output length (default: 8192, max: 32768) |\n\n**Supported formats**: JPEG, PNG, GIF, WebP\n**Size limit**: 10MB\n\n### `understand_audio`\n\nAnalyze audio using Xiaomi MiMo multimodal model.\n\n| Parameter | Type | Required | Description |\n|-----------|------|----------|-------------|\n| `prompt` | string | Yes | Audio understanding task description |\n| `audio_url` | string | No | Single audio URL |\n| `audio_path` | string | No | Single local audio file path |\n| `audio_urls` | list[string] | No | Multiple audio URLs |\n| `audio_paths` | list[string] | No | Multiple local audio file paths |\n| `system_prompt` | string | No | Custom system prompt |\n| `max_tokens` | integer | No | Max output length (default: 8192, max: 32768) |\n\n**Supported formats**: MP3, WAV, FLAC, M4A, OGG\n**Size limit**: URL 100MB, Base64 50MB\n\n### `understand_video`\n\nAnalyze video using Xiaomi MiMo multimodal model.\n\n| Parameter | Type | Required | Description |\n|-----------|------|----------|-------------|\n| `prompt` | string | Yes | Video understanding task description |\n| `video_url` | string | No | Single video URL |\n| `video_path` | string | No | Single local video file path |\n| `video_urls` | list[string] | No | Multiple video URLs |\n| `video_paths` | list[string] | No | Multiple local video file paths |\n| `fps` | float | No | Frames per second, range [0.1, 10], default: 2 |\n| `media_resolution` | string | No | Resolution: \"default\" or \"max\" |\n| `system_prompt` | string | No | Custom system prompt |\n| `max_tokens` | integer | No | Max output length (default: 8192, max: 32768) |\n\n**Supported formats**: MP4, MOV, AVI, WMV\n**Size limit**: URL 300MB, Base64 50MB\n\n## Examples\n\n### Image Understanding\n\n```python\n# URL\nawait understand_image(prompt=\"Describe this image\", image_url=\"https://example.com/image.jpg\")\n\n# Local file\nawait understand_image(prompt=\"What text is in this?\", image_path=\"/path/to/screenshot.png\")\n\n# Multiple images\nawait understand_image(prompt=\"Compare these\", image_urls=[\"url1\", \"url2\"])\n```\n\n### Audio Understanding\n\n```python\n# URL\nawait understand_audio(prompt=\"Transcribe this audio\", audio_url=\"https://example.com/audio.wav\")\n\n# Local file\nawait understand_audio(prompt=\"What is being said?\", audio_path=\"/path/to/audio.mp3\")\n```\n\n### Video Understanding\n\n```python\n# URL with default settings\nawait understand_video(prompt=\"Describe this video\", video_url=\"https://example.com/video.mp4\")\n\n# URL with custom fps and resolution\nawait understand_video(\n prompt=\"Describe the action\",\n video_url=\"https://example.com/video.mp4\",\n fps=5.0,\n media_resolution=\"max\"\n)\n```\n"}}}} |