Rag

Package ai is Hanzo AI — the model API on /v1 (/v1/chat/completions, /v1/messages, /v1/models and the rest of hanzoai/ai's surface) — mounted into a cloud binary with the money, ingest and telemetry callbacks cloud BUILDS but cannot INSTALL.

rag

GET /v1/rag/context

Return every stored chunk of one file_id (full document context). Consolidates the retired chat-rag-api GET /documents/{id}/context.

POST /v1/rag/delete

Delete all chunks of one or more uploaded files (by file_id) from the owner's Search+Vector index. Consolidates the retired chat-rag-api DELETE /documents.

POST /v1/rag/embed

Parse, chunk, and embed one uploaded file under its file_id into the unified Search+Vector index, scoped to the authenticated owner. Provide inline content or a url to fetch+parse (PDF/CSV/XLSX/PPTX/…). Re-embedding the same file_id replaces its chunks. Consolidates the retired chat-rag-api POST /embed and /local/embed.

POST /v1/rag/query-multiple

Retrieve the top-K chunks relevant to a query, scoped to a SET of uploaded files (file_ids). Consolidates the retired chat-rag-api POST /query_multiple. Shares one retrieval path with /rag/query.

POST /v1/rag/query

Retrieve the top-K chunks relevant to a query, scoped to a single uploaded file (file_id). Hybrid keyword+vector retrieval over the same index. Consolidates the retired chat-rag-api POST /query.

---

All Hanzo APIs · Interactive reference