Supabase

Supabase is open-source from day one — its code is publicly available, community-contributed, and can be self-hosted. The company also offers a managed cloud service that packages the same open-source components with infrastructure, scaling, and support. All core products — database, auth, storage, real-time, edge functions — are available in both forms.

Supabase unifies a Postgres database with built-in authentication (with Row Level Security), instant RESTful and GraphQL APIs, real-time data synchronization, edge functions (serverless code), file storage, and vector embeddings for AI/ML workloads. These are usually separate services; Supabase integrates them as one platform.

Supabase includes vector embeddings support that allows developers to store, index, and search embeddings alongside their relational data. It integrates with popular ML models from OpenAI and Hugging Face. This lets users build semantic search, recommendation engines, or retrieval-augmented generation (RAG) applications without leaving the Postgres environment.

Supabase includes built-in authentication with Row Level Security (RLS), which applies database-level access policies per user. It also offers security and performance linters (called Advisors) that scan the project for best practices. Custom domains are supported for Auth endpoints, and the platform uses Postgres's own security model for data protection.

Yes — Supabase provides a local development CLI that lets developers run the entire platform on their machine. Multiple user testimonials highlight the seamless local experience, which mirrors the production cloud environment. Projects can be developed offline and deployed to Supabase's cloud without configuration changes.