Ingest Text

POST

ingest

text

curl --request POST \
  --url https://api.mixpeek.com/v1/ingest/text \
  --header 'Content-Type: application/json' \
  --data '{
  "collection": "col_1234567890",
  "asset_update": {
    "asset_id": "<string>",
    "mode": "replace"
  },
  "metadata": {
    "author": "John Doe",
    "category": "Research Paper",
    "tags": [
      "AI",
      "Machine Learning"
    ]
  },
  "feature_extractors": {
    "embed": [
      {
        "embedding_model": "multimodal",
        "type": "text",
        "value": "lorem ipsum 1"
      },
      {
        "embedding_model": "keyword",
        "type": "text",
        "value": "lorem ipsum 2"
      },
      {
        "embedding_model": "text",
        "type": "text",
        "value": "lorem ipsum 3"
      }
    ],
    "json_output": {
      "response_shape": {
        "key_phrases": [
          "str"
        ],
        "sentiment": "str"
      }
    },
    "entities": {
      "taxonomy_extraction": {
        "taxonomy": "tax_123",
        "assignment": {
          "enabled": false,
          "mode": "threshold",
          "append": false,
          "confidence_threshold": 0.5
        }
      }
    }
  },
  "skip_duplicate": true
}'

{
  "task_id": "<string>",
  "status": "PENDING",
  "inputs": [
    "<any>"
  ],
  "outputs": [
    "<any>"
  ]
}

Headers

Authorization

string | null

Bearer token authentication using your API key. Format: 'Bearer your_api_key'. To get an API key, create an account at mixpeek.com/start and generate a key in your account settings. Example: 'Bearer sk_1234567890abcdef'

X-Namespace

string | null

Optional namespace for data isolation. This can be a namespace name or namespace ID. Example: 'netflix_prod' or 'ns_1234567890'. To create a namespace, use the /namespaces endpoint.

Body

application/json

collection

string

required

Unique identifier for the collection where the processed asset will be stored, can be the collection name or collection ID. If neither exist, the collection will be created.

asset_update

object | null

Controls how processing results are stored - either creating a new asset or updating an existing one.

metadata

object

Additional metadata associated with the file. Can include any key-value pairs relevant to the file.

feature_extractors

object | null

Settings for text processing.

feature_extractors.embed

object[]

List of embedding settings for generating multiple embeddings. field_name's provided are how the raw text will be inserted, if not provided, the field_name will be auto-generated. Default: [{type: 'text', embedding_model: 'multimodal'}] if none provided.

        Request model for embedding generation.
        
        When multiple EmbeddingRequests use the same embedding_model:
        - All inputs will be embedded in the same vector space
        - The final embedding will be the average of all individual embeddings
        - Original values will be stored with a ' | ' separator
        
        Example:
            Two requests with same model "clip":
            1. type: "text", value: "a dog", embedding_model: "clip"
            2. type: "url", value: "https://example.com/image.jpg", embedding_model: "clip"
            
            Result:
            - vectors["clip"] = average of both embeddings
            - embedding_configs["clip"] = "a dog | https://example.com/image.jpg"

feature_extractors.json_output

object | null

Settings for structured JSON output of text analysis.

feature_extractors.entities

object | null

Settings for extracting entities from text content

skip_duplicate

boolean | null

default:

true

Skips processing when a duplicate hash is found and stores an error by the task_id with the existing asset_id

Response

200

application/json

Successful Response

task_id

string

required

status

enum<string>

Available options:

DONE,

FAILED,

SKIPPED,

CANCELLED,

PROCESSING,

DOWNLOADING,

INITIALIZING,

UPLOADING,

QUEUED,

PENDING

inputs

any[] | null

outputs

any[] | null

Was this page helpful?

Suggest edits Raise issue

List Features Ingest Video Url

curl --request POST \
  --url https://api.mixpeek.com/v1/ingest/text \
  --header 'Content-Type: application/json' \
  --data '{
  "collection": "col_1234567890",
  "asset_update": {
    "asset_id": "<string>",
    "mode": "replace"
  },
  "metadata": {
    "author": "John Doe",
    "category": "Research Paper",
    "tags": [
      "AI",
      "Machine Learning"
    ]
  },
  "feature_extractors": {
    "embed": [
      {
        "embedding_model": "multimodal",
        "type": "text",
        "value": "lorem ipsum 1"
      },
      {
        "embedding_model": "keyword",
        "type": "text",
        "value": "lorem ipsum 2"
      },
      {
        "embedding_model": "text",
        "type": "text",
        "value": "lorem ipsum 3"
      }
    ],
    "json_output": {
      "response_shape": {
        "key_phrases": [
          "str"
        ],
        "sentiment": "str"
      }
    },
    "entities": {
      "taxonomy_extraction": {
        "taxonomy": "tax_123",
        "assignment": {
          "enabled": false,
          "mode": "threshold",
          "append": false,
          "confidence_threshold": 0.5
        }
      }
    }
  },
  "skip_duplicate": true
}'

{
  "task_id": "<string>",
  "status": "PENDING",
  "inputs": [
    "<any>"
  ],
  "outputs": [
    "<any>"
  ]
}

Health

Features

Ingest Assets

Entities

Assets

Collections

Namespaces

Feature Extractors

Tasks

Headers

Body

Response