Skip to main content
POST
/
api
/
user
/
embeddings
Get embedding chunks for an upload
curl --request POST \
  --url https://pria.praxislxp.com/api/user/embeddings \
  --header 'Content-Type: application/json' \
  --header 'x-access-token: <api-key>' \
  --data '
{
  "upload": "665a1b2c3d4e5f6789012300"
}
'
{
  "success": true,
  "data": [
    {
      "_id": "665a1b2c3d4e5f6789012345",
      "chunkText": "This is a paragraph from the uploaded document...",
      "chunkLen": 512,
      "chunkIndex": 0,
      "upload": "665a1b2c3d4e5f6789012300",
      "chunkUrl": "#page=2",
      "created": "2023-11-07T05:31:56Z",
      "usage": 128
    }
  ],
  "embeddingModel": {
    "name": "text-embedding-3-small",
    "provider": "openai_cli",
    "maxInputTokens": 8191
  },
  "currentEmbeddingModel": {
    "name": "text-embedding-3-small",
    "provider": "openai_cli"
  },
  "chunkMaxChars": 8000
}

Documentation Index

Fetch the complete documentation index at: https://docs.praxis-ai.com/llms.txt

Use this file to discover all available pages before exploring further.

Authorizations

x-access-token
string
header
required

JWT token passed in x-access-token header

Body

application/json
upload
string
required

The ID of the upload whose embedding chunks to retrieve

Example:

"665a1b2c3d4e5f6789012300"

Response

Embedding chunks retrieved successfully

success
boolean
Example:

true

data
object[]

Embedding chunks sorted by chunkIndex (max 1000)

embeddingModel
object

Model actually used to generate these chunks' vectors. Null when the upload has no recorded embedding model.

currentEmbeddingModel
object

The institution's currently-configured embedding model (what a re-ingestion would use). Compare against embeddingModel to detect drift.

chunkMaxChars
integer

Soft cap (characters) for segment edit/create UI. Matches the server-side RAG chunk-size constant so admin edits stay in lockstep with new ingestion output.

Example:

8000