OrcAI Documentation
Learn how to use OrcAI and run it in your own environment.
OrcAI is a self-hostable platform for building governed AI assistants around curated knowledge. It supports individual, team, education, and research workflows: operators upload content, OrcAI processes and indexes it, and users chat with assistants that can retrieve relevant context from those materials.
Early draft
OrcAI is still in active development and is not yet ready for production use. Some TODOs are intentionally left visible while product details settle.
Open source
OrcAI is licensed under the GNU Affero General Public License, version 3 only (AGPL-3.0-only). The complete licence is available in the repository's LICENSE file.
Start Here
Platform overview
Understand what OrcAI does, who it is for, and how retrieval-backed chat fits together.
Using OrcAI
Follow user-facing workflows for chat, content management, bots, access, and account settings.
Self hosting
Run OrcAI with Docker Compose or manually managed infrastructure.
Development
Work with the repository, commands, architecture notes, and contribution flow.
Main Capabilities
- AI-assisted chat grounded in curated source content
- Document ingestion for PDFs, Office files, and images
- Background processing and indexing workflows
- Qdrant-backed vector search for retrieval-augmented generation
- S3-compatible storage for uploaded and processed assets
- Organization, group, invitation, and resource access management
- Configurable providers, models, quotas, observability, and email delivery