Skip to content

Knowledge base & RAG

Index your documents and answer questions over them with AI. Chunking, retrieval, and GraphRAG — included.

Q&A over your documents

Upload PDFs, DOCX, and text files. Ask questions in natural language. Get cited answers backed by your content.

Multi-format ingestion

PDF, DOCX, TXT, and Markdown. Upload via the dashboard or the ingest-file API. Status tracked per ingestion job.

Smart chunking

Documents are split into semantic chunks with overlap. Chunk size and strategy configurable per knowledge space.

Semantic retrieval

OpenAI embeddings for semantic similarity. BM25 for keyword precision. Hybrid scoring for best-of-both results.

GraphRAG

Entity extraction builds a knowledge graph over your documents. Multi-hop reasoning surfaces connections plain search misses.

Cited answers

Every answer includes source references with document name, chunk position, and relevance score.

Per-tenant spaces

Each org has isolated knowledge spaces. Users only see their own data. No cross-tenant information leakage.

RAG-ready ingestion

Upload files or URLs and let AACsearch prepare searchable, answerable context.

json
// POST /v1/knowledge/:spaceId/ingest-url
{
  "url": "https://docs.example.com/api-reference",
  "options": { "chunk_size": 512, "chunk_overlap": 64 }
}

// POST /v1/knowledge/:spaceId/ask
{
  "question": "How do I authenticate API requests?",
  "context": "api_reference"
}

Launch Search OS in one evening

Create an index, add documents and connect search from your app. Free tier covers a prototype — data persists when you upgrade.

Start freeOpen docs