Infinity ♾️

Infinity serves embeddings, reranking, classification and decision models. Choose the documentation for your server:

• Infinity v3: the native Rust and LibTorch server. Start with the quickstart or browse supported models.
• Python / v2: the existing Python server, including CLI, Python engine and client documentation.

V3 is the direction for future development. Both generations remain documented here while users migrate; choose the matching section when configuring a server or integrating its API.