Storing upserts embedded chunks into a vector database. The collection is created automatically if it doesn’t exist.
rag.store(
embedded_chunks=embed_result.items,
vector_db="qdrant",
collection="docs",
url="http://localhost:6333",
)
Databases
| Database | Extra | Notes |
|---|
| Qdrant | ragrails[qdrant] | Local or cloud. Needs url. |
| Pinecone | ragrails[pinecone] | Cloud. Uses PINECONE_API_KEY. |
| Weaviate | ragrails[weaviate] | Local or cloud. Needs url. |
Keeping an index current
| Task | Method |
|---|
| Add new chunks | store() |
| Replace chunks (re-embeds by ID) | edit() |
| Remove chunks by ID | delete() |
Give chunks stable IDs at chunk time so edit() and delete() can target them later.
Reference
Full parameters: SDK storing · CLI · REST. 
