MCP (Model Context Protocol) server for FG-CLIP embedding services. To obtain and configure the API key, please apply at https://research.360.cn/sass.
This MCP server provides the following tools and resources:
- text_embedding: Generate embedding vectors for text
 - image_embedding: Generate embedding vectors for images
 - cosine_similarity: Compute cosine similarity between two lists of vectors
 
This MCP server helps users achieve the following capabilities:
- Image Feature Extraction: Convert images into high-dimensional vector representations for machine learning and similarity computation
 - Text Feature Extraction: Transform text into semantic vector representations with multi-language support
 - Multi-modal Similarity Computation:
- Image-to-Image Similarity: Compare visual similarity between different images
 - Image-to-Text Similarity: Enable cross-modal retrieval, such as finding relevant images based on text descriptions
 - Text-to-Text Similarity: Calculate semantic similarity between texts
 
 
Through these capabilities, users can build powerful search engines, recommendation systems, content classification, and multi-modal AI applications.
Generate embedding vectors for input texts.
Parameters:
texts: A list of text strings to embedmodel: The model to use (default: "fg-clip")
Returns:
saved_uris: A list of URIs where the embeddings are storedsuccess: Whether the operation succeedederror_msg: Error message, if any
Generate embedding vectors for images.
Parameters:
images: A list of image URLs or base64-encoded imagesmodel: The model to use (default: "fg-clip")
Returns:
saved_uris: A list of URIs where the embeddings are storedsuccess: Whether the operation succeedederror_msg: Error message, if any
Compute cosine similarity between two lists of vectors.
Parameters:
uris_a: A list of URIs for the first set of embeddingsuris_b: A list of URIs for the second set of embeddingsmode: Calculation mode (default: "pairwise")"pairwise": Compute similarity for vectors at corresponding positions"matrix": Compute a full similarity matrix for all vector pairs
Returns:
similarities: Similarity values or a similarity matrixshape: Shape information of the resultsuccess: Whether the operation succeeded
git clone https://github.com/360CVGroup/FGCLIP-MCP 
cd FGCLIP-MCP
uv venv
uv sync
source .venv/bin/activate
export MCP_API_KEY=your_api_key 
pytest -q{
  "mcpServers": {  
    "fgclip-mcp": {
      "command": "uvx",
      "args": [
        "fgclip-mcp"
      ],
      "env": {
        "MCP_API_KEY": "your_api_key"
      }
    }
  }
}{
  "mcpServers": {  
    "fgclip-mcp-local": {
      "command": "uv",
      "args": [
        "--directory",
        "/path_to_fgclip-mcp/src/fgclip_mcp",
        "run",
        "/path_to_fgclip-mcp/src/fgclip_mcp/__main__.py"
      ],
      "env": {
        "MCP_API_KEY": "your_api_key"
      }    
    }
  }
}
Use Case in Cursor IDE
Chat with MCP
Example: Searching for images based on given text

Image URLs:
- https://p0.qhimg.com/t11098f6bcd000b4fb05d7bf627.jpg
 - https://p0.qhimg.com/t11098f6bcdc3c5f3e99a1dbfad.jpg
 
Apache License 2.0




