Infino AI

Infino is an open-source embedded retrieval library that runs full-text, vector, hybrid and SQL queries over one table stored as ordinary Parquet files on local disk or object storage, with a hosted option for agents.

Why it matters

For teams building AI agents that need search, semantic lookup and analytics over the same data. Instead of syncing a search engine, a vector database and a warehouse, it keeps one Parquet copy in your own bucket.

Highlights

• One table supports BM25 full-text, vector and hybrid search with fused ranking, plus SQL.
• SQL functions such as hybrid_search and graph_walk let an agent combine keyword, semantic and graph lookups with joins and GROUP BY in one query.
• Installs as a Python package (pip install infino), a Node.js package or a Rust crate, and stores data as plain Parquet locally or on object storage.
• The README reports warm p50 latency of 591 μs for vector top-10 and 125 μs for BM25 top-10 at 1 million documents on object storage, measured in its CI on Azure Blob.unverified

Taken from the project's own README, docs or launch page. Claims about speed or accuracy are the maker's unless stated otherwise.

Worth knowing: The library installs the mimalloc allocator by default; turn that feature off if your process already sets a global allocator.

Quick facts

Links

• Official site: https://infino.ai/
• Repository: https://github.com/infino-ai/infino
• Release notes: https://github.com/infino-ai/infino/releases
• Docs: https://infino.ai/docs

Related ranking

See how established tools compare in Best AI Agents That Do Tasks for You. New releases aren't scored until we review them against the rubric.

What we couldn't verify

• quick_facts.pricing (hosted plan rates)
• highlights[3] (the maker's own benchmarks)

Checked 2026-11-03 from the tool's own pages. Prices and star counts change, so confirm on the official site. Listing here is not a ranking or an endorsement.