How it works
Four steps from raw content to production-ready knowledge base.
Ingest
Pull content from Confluence, web URLs, file uploads, or paste text directly.
Curate
AI chunks your content, then you review, edit, split, and merge with full control.
Publish
Preview, validate, then atomically publish to your vector collection.
Retrieve
Search your knowledge base or chat with it using RAG-powered answers and citations.
Everything you need for production RAG
From ingestion to retrieval — a complete platform for building and maintaining knowledge bases.
Multi-Source Ingestion
Pull from Confluence, web URLs, manual text, or upload PDF, DOCX, Markdown, and CSV files directly.
Session-Based Curation
Draft sessions let you review, edit, split, merge, and reorder chunks before publishing. Human in the loop.
AI Quality Scoring
AI assistant analyzes chunk quality, suggests operations, and scores with an approval flow.
Configurable Chunking
Choose between LLM semantic chunking or fast character-based splitting with size and overlap control.
Chat Playground
Ask questions against your knowledge base. RAG-powered answers grounded in your chunks with cited sources.
Atomic Publish
Preview, validate, then atomically replace collection contents. Zero-downtime knowledge updates.
Built for production
A composable architecture with battle-tested infrastructure.
Open source. Self-hosted. Yours.
RAGler runs on your infrastructure. No vendor lock-in, no data leaving your network. Deploy with Docker Compose in minutes.