incredlabs

Frequently asked questions

What incredlabs builds, how it's licensed, and how ORMDB and Liath differ.

What is incredlabs?
incredlabs is an open-source software company building Rust-native, high-performance data infrastructure for AI — including ORMDB, a database that speaks ORM natively, and Liath, the SQLite for AI agents.
What products does incredlabs make?
The public products are ORMDB (a relational database that fetches object graphs natively) and Liath (programmable memory for AI agents, queried with sandboxed Lua). Additional data-platform components are in development.
Are incredlabs products free and open source?
Yes. incredlabs products are open source under the MIT License and developed in the open at github.com/incredlabs. There is no license fee.
What language are incredlabs products written in?
The cores are written in Rust for performance, memory safety, and portability. Client libraries and bindings are available in TypeScript, Python, and other languages depending on the product.
How is ORMDB different from Liath?
ORMDB is a relational database that speaks ORM natively — ideal for application backends that need graph fetches, relations, and safe migrations. Liath is programmable memory for AI agents — an embedded store where agents write Lua to query their own memory. Use ORMDB for your app data and Liath for agent memory.
What is Liath, and how is it different from a vector database?
Liath is the SQLite for AI agents: an embedded memory store where the agent writes sandboxed Lua to query its own memory, instead of calling a fixed vector-search API. A vector database exposes a single similarity call and a service to operate; Liath gives the agent a real programming language to filter, re-rank, and cross-reference results, and runs embedded with zero infrastructure. Reach for a vector DB when you need a shared, horizontally scaled index; reach for Liath when you want per-agent, programmable memory with a hard security boundary and no server.
How is ORMDB different from Postgres plus a traditional ORM?
A traditional stack pairs a relational database with an application-layer ORM (Prisma, Drizzle, TypeORM, SQLAlchemy) that translates object graphs into SQL — the source of N+1 queries and object-relational impedance mismatch. ORMDB moves that graph awareness into the database, so relations, safe online migrations, and vector, geo, and full-text search are native rather than bolted on with extensions and extra services. ORMDB also ships drop-in adapters for those same ORMs.
How do I get started with incredlabs products?
Install ORMDB with `cargo install ormdb-server` or Liath with `pip install liath` (or the `liath-rs` crate for Rust). Then follow the guides at docs.incredlabs.com. Both products are open source on GitHub at github.com/incredlabs, so you can read the code and run everything locally.
Was incredlabs previously called something else?
Yes. incredlabs was formerly known as Skelf Research. The products, mission, and open-source approach are unchanged.
Where can I find the documentation?
Documentation for all products lives at docs.incredlabs.com. Each product also links to its docs from its website and GitHub repository.
How do I contact incredlabs?
Email contact@incredlabs.com or open an issue on GitHub at github.com/incredlabs.

Still have questions? Dig into ORMDB and Liath, read about incredlabs, or get in touch.