Faiss

license

MIT(OSI)

source

public

core features withheld

no

MIT license body confirmed. The repository is public and unarchived and builds the whole product, and the README describes no paid tier, enterprise edition or license-gated build beside it, so source is public and the core ungated.

• https://raw.githubusercontent.com/facebookresearch/faiss/main/LICENSE recorded 2026-09-30

  LICENSE file is the MIT text: "MIT License" and "Permission is hereby granted, free of charge, to any person obtaining a copy".

• https://raw.githubusercontent.com/facebookresearch/faiss/main/README.md recorded 2026-09-30

  README describes a library for "similarity search and clustering of dense vectors" that is "written in C++ with complete wrappers for Python/numpy" and lists "the only dependency" as a BLAS implementation with optional GPU support; searched the body for enterprise, paid, license key and commercial: none.

`faiss-cpu` is a community-maintained wheel of the library rather than a client for a separate service, so its PyPI downloads are downloads of Faiss itself. The vendor's own distribution channel is conda, which publishes no comparable download figure, so this understates rather than overstates total use.

• https://pypistats.org/api/packages/faiss-cpu/recent recorded 2026-08-18

  last_month downloads = 15,698,726 for faiss-cpu

Faiss offers the broadest index family in the category, spanning flat, IVF, PQ, OPQ, HNSW and composite indexes with GPU support. It has no persistence service, filtering layer or serving surface of its own, which is why it covers much less of the persist-index-retrieve path than Vespa does.

• https://github.com/facebookresearch/faiss/blob/main/README.md recorded 2026-08-18

  README still documents a library for similarity search and clustering of dense vectors with algorithms that scale past RAM, written in C++ with complete Python wrappers and GPU implementations of the most useful algorithms, developed at Meta's Fundamental AI Research group.