geldata/gel

Health v2 · maintenance only

Adoption not part of the score

14171 stars · 452 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Gel is an open-source graph-relational database built on top of Postgres, combining a strict object-type schema, the EdgeQL query language, and built-in auth, vector search, and AI/RAG features. It ships with a CLI, graphical UI, migrations engine, and typed client SDKs for Python and TypeScript, plus a managed cloud offering.

Use cases

• model application data as object types instead of tables
• replace SQL and ORMs with a type-safe query language
• add OAuth and password login to an app without a separate auth service
• store embeddings and run RAG over my data
• run hierarchical graph-like queries without JOINs
• manage schema migrations with a modern database
• build a Next.js or FastAPI app with a typed data layer

When to choose

• you want Postgres reliability with an object-oriented schema and composable queries
• you need integrated auth, vector search, and AI features in one data platform
• you value end-to-end type safety with generated TypeScript or Python clients
• you want a zero-config local dev experience with easy cloud deployment

When to avoid

• your team and tooling are deeply invested in standard SQL and existing Postgres workflows
• you need a database with a long track record and broad third-party ecosystem support
• you require exotic Postgres extensions or low-level tuning not exposed by Gel
• you cannot adopt a proprietary query language (EdgeQL) alongside SQL

Facets

service · maturity active

database orm auth rag vector-database search-engine cli sdk databases backend web-development developer-tools large-language-models windows self-hosted cloud python edgeql graph-relational postgres schema-migrations type-safety code-generation oauth embeddings linux macos docker nodejs web-server

Member repositories

For agents

markdown · JSON · MCP: product_card(name="geldata/gel")