Korvus
One stop RAG Search SDK
- Postgres Native RAG: Perform complex RAG operations directly in the database without the need for external services and API calls.
- Single Query Efficiency: The entire RAG process, from embedding generation to text generation, is executed in one SQL query.
- Scalability and Performance: Built on Postgres, it inherits its excellent scalability and performance features.
- Simplified Architecture: Replace complex service-oriented architecture with a single powerful query.
- High Performance: Eliminate API calls and data movement to achieve faster processing and higher reliability.
- Open Source: Open source software and models that provide a local Docker runtime experience.
- Multi Language Support: Supports multiple programming languages such as Python, JavaScript, and Rust.
- Unified Pipeline: Combining embedding generation, vector search, reordering, and text generation in a single query.
Product Details
Korvus is a search SDK built on Postgres, which unifies the entire RAG (Retrieval Enhanced Generation) process into a single database query. It provides high-performance, customizable search capabilities while minimizing infrastructure considerations. Korvus utilizes the pgml and pgvector extensions of PostgresML to compress the RAG process within Postgres. It supports multiple language SDKs, including Python, JavaScript, Rust, and C, allowing developers to seamlessly integrate into existing technology stacks.
