Search and retrieve documents using TF-IDF and cosine similarity in memory. Built for developers who need lightweight vector search without external dependencies.
io.github.Br0ski777/vector-search
https://vector-search-x402-production.up.railway.app/mcp
SSE
No auth required
How models use it and what it is built for.
Search and retrieve documents using TF-IDF and cosine similarity in memory. Built for developers who need lightweight vector search without external dependencies.
Hosted endpoint — paste into any MCP client.
Where to find authoritative docs and source for Vector Search API.
Paste any of these into Agent Studio after connecting Vector Search API.
Common questions about connecting and running Vector Search API.
What vector search algorithms does this MCP server use?
The server uses TF-IDF (term frequency-inverse document frequency) and cosine similarity for in-memory vector search. These are lightweight algorithms suitable for moderate-sized document collections without requiring external ML libraries.
Is data persisted or stored only in memory?
The server stores vectors in memory, so data is not persisted across restarts. This design prioritizes speed and simplicity for development and testing workflows.
How do I authenticate and what is the x402 micropayment?
The registry metadata mentions x402 micropayment support, indicating the server may use a micropayment protocol for usage. Consult the hosted endpoint documentation or contact the maintainer for authentication and payment setup details.
What are the limits on document size or collection size?
Since this is an in-memory implementation, performance depends on available RAM. The registry does not specify hard limits; test with your expected document volume to determine practical constraints.
Can I use this for production or is it development-only?
The in-memory design and lack of persistence make this best suited for development, prototyping, and testing. For production use cases requiring durability and scale, consider a dedicated vector database like Pinecone or Weaviate.
Connect Vector Search API to Claude, GPT, Gemini, DeepSeek and 30+ AI models in MCP Agent Studio. Compare answers side-by-side, save reusable agent presets, share runs — all in your browser, no install required.
Open Agent Studio