Lower memory footprint · facebookresearch/faiss Wiki
The indexes we have seen, IndexFlatL2 and IndexIVFFlat both store the full vectors. To scale up to very large datasets, Faiss offers variants that compress the stored vectors with a lossy compression based on product quantizers.
Lower memory footprint · facebookresearch/faiss Wiki · GitHub Skip to content You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert {{ message }} Uh oh! There was an error while loading. Please reload this page . facebookresearch / faiss Public Notifications You must be signed in to change notification settings Fork 4.5k Star 40.6k Lower memory footprint Jump to bottom Gergely Szilvasy edited this page Sep 29,
Explore this link on the map →related reading
- Faiss indexes · facebookresearch/faiss Wiki · GitHubgithub.com
- Faiss: A library for efficient similarity search - Engineering at Metaengineering.fb.com
- Product Quantization: Compressing high-dimensional vectors by 97% | Pineconepinecone.io
- Indexing 1T vectors · facebookresearch/faiss Wiki · GitHubgithub.com
- TurboQuant: Redefining AI efficiency with extreme compressionresearch.google
- Welcome to Faiss Documentation - Faiss documentationfaiss.ai
- Guidelines to choose an index · facebookresearch/faiss Wiki · GitHubgithub.com
- TurboQuant: Redefining AI efficiency with extreme compressionresearch.google
- Getting started · facebookresearch/faiss Wiki · GitHubgithub.com
- Faster search · facebookresearch/faiss Wiki · GitHubgithub.com
- GitHub - matsui528/faiss_tips: Some useful tips for faiss · GitHubgithub.com
- Indexes that do not fit in RAM · facebookresearch/faiss Wiki · GitHubgithub.com