Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Vision MCP Server extract text from this image: receipt.jpg"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Give your AI agent eyes. An MCP server providing multimodal vision capabilities โ image analysis, OCR, image comparison, and video analysis โ powered by any OpenAI-compatible vision model.
่ฎฉไฝ ็ AI ไปฃ็ๆฅๆ่ง่ง่ฝๅใ ้่ฟไปปไฝ OpenAI ๅ ผๅฎน็่ง่งๆจกๅ๏ผๆไพๅพๅๅๆใOCR ๆๅญ่ฏๅซใๅพๅๅฏนๆฏๅ่ง้ขๅๆใ
Features ยท Quick Start ยท Tools ยท Models ยท ไธญๆ่ฏดๆ
Tool | Description |
๐ | Analyze images with natural language prompts |
๐ | Extract text from images (plain text / Markdown / JSON) |
๐ | Compare 2โ4 images side by side |
๐ฌ | Analyze video content (requires video-capable model) |
Plus:
๐ OpenAI-compatible โ Works with any vision model via standard API
๐ Local files & URLs โ Auto-converts local files to base64
โ๏ธ Configurable โ Environment variables, config files, or both
Related MCP server: vision-mcp
git clone https://github.com/Loveacup/vision-mcp-server.git
cd vision-mcp-server
npm install && npm run buildCreate a .env file in the project root:
VISION_BASE_URL=http://your-server:port/v1/chat/completions
VISION_MODEL=Qwen3-VL-32B
VISION_API_KEY=your-api-key # optional for local models{
"baseUrl": "http://your-server:port/v1/chat/completions",
"model": "Qwen3-VL-32B",
"apiKey": "your-api-key",
"maxTokens": 4096,
"temperature": 0.7
}npm startThe server communicates over stdio, designed to be launched by an MCP client such as Claude Code.
Add to your ~/.mcp.json:
{
"mcpServers": {
"vision": {
"command": "node",
"args": ["/path/to/vision-mcp-server/dist/index.js"],
"env": {
"VISION_BASE_URL": "http://your-server:port/v1/chat/completions",
"VISION_MODEL": "Qwen3-VL-32B",
"VISION_API_KEY": "your-api-key"
}
}
}
}Replace
/path/to/vision-mcp-serverwith the actual install path.
Configuration priority: environment variables > config file > defaults
Variable | Config Key | Default | Description |
|
| (required) | OpenAI-compatible chat completions endpoint |
|
|
| Model name |
|
| (empty) | API key (optional for local models) |
|
|
| Max response tokens |
|
|
| Sampling temperature |
analyze_imageAnalyze an image with a vision language model.
Parameter | Type | Required | Default | Description |
| string | โ | โ | Local file path or URL |
| string |
| Analysis prompt | |
|
|
| Detail level |
ocr_imageExtract text from an image using OCR.
Parameter | Type | Required | Default | Description |
| string | โ | โ | Local file path or URL |
| string |
| Language hint, e.g. | |
|
|
| Output format |
compare_imagesCompare 2โ4 images and describe differences/similarities.
Parameter | Type | Required | Default | Description |
| string[] | โ | โ | 2โ4 image sources |
| string |
| Comparison prompt |
analyze_videoAnalyze video content. Requires a model with video support (e.g., Qwen3-VL).
Parameter | Type | Required | Default | Description |
| string | โ | โ | Local file path or URL |
| string |
| Analysis prompt |
Model | Provider | Image | Video | Notes |
Qwen3-VL | Self-hosted / API | โ | โ | Recommended. Full multimodal support |
GPT-4o | OpenAI | โ | โ | Strong image analysis |
LLaVA | Self-hosted | โ | โ | Open-source alternative |
InternVL | Self-hosted | โ | โ ๏ธ | Strong multilingual OCR |
Any model served via vLLM, Ollama, LMDeploy, or other OpenAI-compatible servers should work.
Supported formats: JPEG, PNG, GIF, WebP, BMP, SVG | MP4, AVI, MOV, MKV, WebM
vision-mcp-server/
โโโ src/
โ โโโ index.ts # MCP server entry point
โ โโโ config.ts # Configuration loader
โ โโโ types.ts # TypeScript type definitions
โ โโโ tools/
โ โ โโโ analyze-image.ts
โ โ โโโ ocr-image.ts
โ โ โโโ compare-images.ts
โ โ โโโ analyze-video.ts
โ โโโ utils/
โ โโโ api-client.ts # OpenAI-compatible API client
โ โโโ file-handler.ts # Local file โ base64
โโโ package.json
โโโ tsconfig.json
โโโ .env.example
โโโ LICENSEanalyze_image โ ไฝฟ็จ่ง่ง่ฏญ่จๆจกๅๅๆๅพๅ๏ผๆฏๆ่ช็ถ่ฏญ่จๆ้ฎ
ocr_image โ OCR ๆๅญ่ฏๅซ๏ผๆฏๆ็บฏๆๆฌใMarkdownใJSON ่พๅบ
compare_images โ ๅฏนๆฏ 2โ4 ๅผ ๅพๅ๏ผ่ฏๅซๅทฎๅผๅ็ธไผผไนๅค
analyze_video โ ๅๆ่ง้ขๅ
ๅฎน๏ผ้่ฆ Qwen3-VL ็ญๆฏๆ่ง้ข็ๆจกๅ๏ผ
git clone https://github.com/Loveacup/vision-mcp-server.git
cd vision-mcp-server
npm install && npm run build้
็ฝฎ .env๏ผ
VISION_BASE_URL=http://your-server:port/v1/chat/completions
VISION_MODEL=Qwen3-VL-32B
VISION_API_KEY=your-api-keyๅจ Claude Code ็ ~/.mcp.json ไธญๆทปๅ ๏ผ
{
"mcpServers": {
"vision": {
"command": "node",
"args": ["/path/to/vision-mcp-server/dist/index.js"],
"env": {
"VISION_BASE_URL": "http://your-server:port/v1/chat/completions",
"VISION_MODEL": "Qwen3-VL-32B",
"VISION_API_KEY": "your-api-key"
}
}
}
}ๅฐ /path/to/vision-mcp-server ๆฟๆขไธบๅฎ้
ๅฎ่ฃ
่ทฏๅพใ
Analyze an image using a vision language model. Supports local file paths and URLs.
Inputs: image (required), detail, prompt.
Analyze video content using a vision language model. Requires a model with video support (e.g., Qwen3-VL).
Inputs: video (required), prompt.
Compare 2-4 images and describe differences/similarities. Supports local file paths and URLs.
Inputs: images (required), prompt.
Extract text from an image using OCR. Supports plain text, Markdown, and JSON output formats.
Inputs: image (required), format, languages.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelogโฆ
Scored across 4 tools
Each tool targets a distinct task: image content analysis, video analysis, image comparison, and text extraction. There is no overlap in purpose, making selection clear.
All tools follow a consistent verb_noun pattern (analyze_image, analyze_video, compare_images, ocr_image), with no mixing of styles.
With 4 tools, the set is focused and not overwhelming. It covers core vision tasks, though a few more (e.g., image generation) could be added for broader scope.
The tools cover image/video analysis, OCR, and comparison. Minor gaps exist (e.g., image metadata extraction), but the surface is sufficient for common vision use cases.
Image & PDF tools for AI agents: compress, convert, resize, PDF, AI vision, pipeline.
Video, audio, and image processing for AI agents: convert, transcribe, upscale - 150+ operations.
OCR, transcription, file extraction, and image generation for AI agents via MCP.
Generate images, GIFs, videos, and PDFs from HTML, URLs, or templates โ from your AI agent.
