Graph-Relational Database Platform

Gel is a scalable, integrated data platform built on top of Postgres that supercharges the relational model with a modern graph-relational data model, eliminating N+1 query problems, simplifying migrations, and streamlining the entire database workflow. Formerly known as EdgeDB, the project recently joined Vercel and rebranded to Gel. The platform introduces object types with links instead of foreign key joins, enabling hierarchical selects, inserts, and updates through link traversal without JOIN operations. The EdgeQL query language provides compositional, strictly typed queries that outperform equivalent SQL in both readability and performance, while classic SQL and ORM support remain available for compatibility. End-to-end type safety flows from schema definitions through code generation, producing typed query builders for TypeScript and Python, with client libraries also available for Go, Dart, Rust, .NET, and Java. Built-in authentication provides OAuth integrations, email and password, magic links, and WebAuthn passkeys with a production-ready login UI at no additional cost. Artificial intelligence integration includes a vector store, configurable automatic embeddings generation, and a ready-to-use retrieval-augmented generation endpoint that connects to OpenAI, Anthropic, and Mistral through a few schema declarations. Schema migrations are handled through a complete engine with branching and rapid prototyping support. The graphical UI provides a query editor with visual analysis output, schema viewer with graph representation, data explorer with drill-down navigation, and query performance statistics with execution time histograms. Gel Cloud integrates with Vercel and GitHub for automatic preview deployments, with encryption at rest, backups, and monitoring. The platform is fully open source with 14,000 GitHub stars and offers one-click export to vanilla Postgres if needed.