The first month of the year was very productive here at Supabase. Here is a highlight of what we shipped during January:
Storing OpenAI embeddings in Postgres with pgvector
pgvector is a popular PostgreSQL extension for storing embeddings and performing vector similarity search. It was one of the most requested extensions by the AI/ML community and is now available thanks to gregnr.
Meet Supabase Clippy: ChatGPT for Docs
Greg wasted no time and took pgvector for a spin, he combined it with OpenAI to build Supabase Clippy, a next-generation doc search. The first implementation is a 1-week MVP and fully open source, so you can build on top of it.
Client library reference: Python and C#
We have released extensive reference docs for C# and Python, detailing every object and method. What are you going to build?
pg_graphql now supports Views, Materialized Views, and Foreign Tables
Views, Materialized Views, and Foreign Tables are three database objects that provide a powerful way to access and organize and transform data without duplication.
Automatic WebP detection for Image Transformation
WebP is a modern image format that provides superior lossless and lossy compression for images on the web. We are enabling format conversion by default for anyone who has Image Transformations. You can opt out by including format: origin in the transformation parameters.
Quick product updates
- Postgres Extension: Another powerful and time-tested extension, pg_repack, is added to Supabase. [PR] - Auth: Multi-tab session support using the new browser BroadcastChannel API. If a user logs out on one tab, they will now be logged out on all tabs. [PR] - Postgres: Superior speed with lz4 database compression. [PR] - Postgres: Use ICU locales and collations for text attribute ordering in database queries. [PR] - Docs: New guide on scheduling functions with pg_cron. [Guide] - Edge Functions: You can now download source codes of deployed edge functions from the CLI. [Doc]