Infinity

Infinity is an AI-native database designed for LLM (Large Language Model) applications, offering fast and flexible hybrid search across various data types.

Features

• Hybrid Search: Supports fast search across dense vectors, sparse vectors, tensors (multi-vectors), and full-text data.
• Rich Data Types: Handles strings, numerics, vectors, tensors, and structured data.
• LLM Application Support: Suitable for use cases such as search, recommenders, question answering, conversational AI, copilots, content generation, and retrieval-augmented generation (RAG) applications.
• Performance: Designed for high performance and scalability; benchmark results available.
• Flexible Deployment:
  • Embedded Mode: Can be embedded directly into Python applications without a separate backend server.
  • Client-Server Mode: Supports deployment as a separate server and client process.
• Ease of Use: Provides a Python API and documentation for quick integration and usage.
• Open Source: Source code available on GitHub.

Pricing

No pricing information is provided; the project appears to be open source.

Links

• GitHub Repository