Create a Database

Databases are the heart of our Retrieval-Augmented Generation (RAG) system.

Document Databases store and serve the source materials that knowledge workflows rely on: files, email contents, and structured documents. In this stack, documents are linked to conversations (data rooms) and Insights via short-lived references, enabling retrieval, processing, and indexing while preserving provenance and access controls.

Each database can have:

• Folders: Can be used for organization by project, topic, or knowledge domain. Fine-granular access can be provided to specific users.

• Documents: Main entities of different types like PDFs, Word/Office docs, emails, spreadsheets, structured JSON/XML, images that will be indexed, chunked and vectorized for RAG.

  • Documents can be uploaded manually, by API or imported through the provided web crawler.

Create a new Knowledgebase / Database

Create Knowledgebase

post

Authorizations

HTTPBearer

AuthorizationstringRequired

Bearer authentication header of the form Bearer <token>.

Body

application/json

namestringRequired

descriptionany ofRequired

stringOptional

or

nullOptional

is_publicbooleanOptionalDefault: false

embeddingModelstring · enumOptionalDefault: azure-ada-002Possible values:

chunkSizeany ofOptionalDefault: 1200

integer · max: 30000Optional

or

nullOptional

chunkOverlapany ofOptionalDefault: 200

integer · max: 12000Optional

or

nullOptional

markedLanguagesstring[] · max: 3OptionalDefault: []

enableExtractTableany ofOptionalDefault: false

booleanOptional

or

nullOptional

enableExtractImageany ofOptionalDefault: true

booleanOptional

or

nullOptional

enableConvertImageany ofOptionalDefault: false

booleanOptional

or

nullOptional

useContextualChunkingany ofOptionalDefault: false

booleanOptional

or

nullOptional

enableMultiplePdfPagesPerChunkbooleanOptionalDefault: false

Responses

200

Successful Response

application/json

422

Validation Error

application/json

post

/knowledge_base

HTTPcURLJavaScriptPython

POST /knowledge_base HTTP/1.1
Authorization: Bearer YOUR_SECRET_TOKEN
Content-Type: application/json
Accept: */*
Content-Length: 299

{
  "name": "text",
  "description": "text",
  "is_public": false,
  "embeddingModel": "azure-ada-002",
  "chunkSize": 1200,
  "chunkOverlap": 200,
  "markedLanguages": [
    "text"
  ],
  "enableExtractTable": false,
  "enableExtractImage": true,
  "enableConvertImage": false,
  "useContextualChunking": false,
  "enableMultiplePdfPagesPerChunk": false
}

{
  "code": 1,
  "key": "text",
  "body": "text"
}

The response contains a body property that contains the _id of the created Knowledgebase. This id will can be used for altering the DB, creating folders and uploading documents.

Next Step

Now that you have created a Knowledgebase, you can Create Folders and Upload Documents to DBs.

Supported File-Types

Every standard file-type is already being supported by our indexing and vectorization pipelines. Still, we are constantly adding more compatibility.