Infinity (InfiniFlow)

Infinity is a search database for developers building retrieval-augmented generation (RAG), question answering and recommendation systems. It lets applications combine matches based on meaning with exact text matches and structured filters.

• Hybrid retrieval: Search dense embeddings, sparse embeddings, multi-vector tensors and full text together, with filters to narrow the results.
• Result ranking: Combine or rerank matches using reciprocal rank fusion (RRF), weighted sums or ColBERT.
• Structured records: Store strings, numbers and vectors in the same database so retrieval can use both embeddings and other record fields.
• Application access: Python and HTTP APIs connect applications to the database, and an embedded Python module supports retrieval within a Python application.

Infinity is open source under Apache 2.0. You can self-host it as a separate server using Docker or a standalone binary, or use the embedded module. Its single-binary architecture has no dependencies.

Supported environments include Linux and macOS, plus Windows through WSL or WSL2. The stated CPU requirement is x86_64 with AVX2 support; Python use requires Python 3.11 or later.