⚠️ Pre-alpha: This platform is in early development. Features may change.

Embedding API for semantic search, RAG, recommendations & multimodal vectors — raggen.ai

Infrastructure for AI devs
raggen.ai

Build RAG features, not plumbing

Multimodal vector search and Gemini embeddings—upload once, query by meaning, subscribe to live index updates.

Upload images, video, audio, and documents once. Search by embedding similarity across bases—GraphQL queries, subscriptions, and MCP included.

Everyday embedding jobs

Simplify your common tasks

Five jobs teams reach for every day—search, Q&A, recommendations, deduping uploads, and mixed-media findability.

raggen.ai handles embeddings, storage, and search so you can focus on the product—not tags, tuners, or glue code.

One platform

Built in, not bolted on

Upload, embed, query, and subscribe from one place—so you ship sooner and spend less time wiring vendors.

Embedding models

Vector dimensions

GraphQL subscriptions

Supported media

Similarity search, ready to call

Run a topK vector query in GraphQL and get ranked results back—no custom search service to deploy first.

topK · 768d

query FilesWithinDistance(
  $vector: [Float!!]
  $topK: Int!
  $filter: EmbeddingIndexFilter!
) {
  querySimilarEmbeddingIndexByEmbedding(
    vector: $vector
    topK: $topK
    by: embedding
    filter: $filter
  ) {
    id
    model
    vector_distance
    file { id name }
  }
}

GE2-768

768d · Gemini

GE2-1536

1536d · Gemini

Your files, organized

  • Private org workspaces—no accidental cross-customer leaks
  • Group content into bases the way your product thinks
  • Text extracted automatically for search and RAG
  • One upload path for every supported media type

Search that stays fresh

  • Similarity search with distance scores you can trust
  • Pick 768, 1536, or 3072 dimensions when you need more accuracy
  • Indexes update live—no nightly rebuild jobs
  • Same API for quick search queries and corpus files

Plug into your stack

  • GraphQL queries and subscriptions from any client
  • API keys for backend jobs and CI
  • MCP for Cursor, Claude Desktop, and custom tools
  • Interactive docs and copy-paste examples
MCP integration

Use your files inside your editor

Connect Cursor or Claude Desktop once—ask questions over your bases without leaving the app you're already in.

Cursor

Add raggen as an MCP server and pull context into the editor while you code.

Claude Desktop

Query org bases and files from Claude over the MCP transport.

Custom MCP clients

Any MCP-compatible client can call the same GraphQL-backed tool surface.

https://mcp.raggen.aiAPI & MCP docs
In production

Our customers

Teams shipping search and RAG products on raggen.ai.

Pricing

Start on the free tier

No credit card required—monthly token and storage allowance included on every org.

No credit card required

Free tier

Included monthly allowance · overage billed only if exceeded

Start on the free tier: 1 million embedding tokens per org each month and 100 MB storage included. Overage at $0.40/1M tokens and $0.05/GB-month only when you exceed the allowance.

Free tier tokens
1 million / mo
Free tier storage
100 MB
Token overage
$0.40 / 1M
Storage overage
$0.05 / GB-mo
  • No credit card to start
  • Monthly token allowance included
  • Usage breakdown in Account
  • One invoice per org per month (overage only)
  • Prepaid balance applied before invoice due

Free tier first; overage only if you exceed the allowance

View free tier details
FAQ

Questions, answered

Quick answers about files, models, search, and getting started.

Get started

Ready to try it?

Create a base, upload a few files, and run your first similarity search—free tier included.