Infino

Infino is a data infrastructure designed for developers. The problem it aims to solve is that when AI agents need to search through full texts, vectors, and SQL data simultaneously, it often requires the use of multiple separate systems. Infino integrates these capabilities into the same Parquet dataset stored in object storage, eliminating the need for redundant data synchronization.

The official website publishes the results of public benchmarks such as ClickBench and VectorDBBench. It also provides comparison pages showing how the product performs compared to tools like Elasticsearch, Qdrant, Pinecone, and ClickHouse, and claims that its cost is approximately one-tenth of that of traditional solutions. However, these figures are mainly based on internal tests conducted by the vendor; it is advisable to run tests using your own data before deploying the product in a production environment.

The project is open-sourced on GitHub and comes with free introductory materials. No credit card is required for participation. The team has just received $7.5 million in funding from Bessemer. This project is not suitable for ordinary users; it is more geared towards backend engineers working on proxy products.