Tensorlake Datasets API

openapi: 3.1.0
info:
  title: Tensorlake Datasets API
  description: 'Tensorlake Cloud APIs for Document Ingestion, Serverless Workflows, and Sandboxes. The Document Ingestion API parses documents (PDF, images, office formats) into layout-aware Markdown and structured chunks, performs schema-guided structured extraction and classification, and manages reusable files and datasets. Parsing runs as an asynchronous job: a parse request returns a parse id (job id) that is polled for results or delivered via webhook. All requests are authenticated with an API key passed as a Bearer token (Authorization: Bearer tl_apiKey_...).

    GROUNDING NOTE (API Evangelist, 2026-07-12): The path list, HTTP methods, base URL, and Bearer auth are grounded in Tensorlake''s published OpenAPI document (docs.tensorlake.ai/api-reference/openapi.yaml) and API reference. Request and response BODY SCHEMAS below are MODELED from Tensorlake documentation and SDK behavior and are illustrative, not byte-exact - verify field names against the live specification before generating client code.'
  version: 0.1.0
  contact:
    name: Tensorlake
    url: https://www.tensorlake.ai
servers:
- url: https://api.tensorlake.ai
  description: Tensorlake Cloud
security:
- bearerAuth: []
tags:
- name: Datasets
  description: Group documents under a shared parse/extraction configuration.
paths:
  /documents/v2/datasets:
    post:
      operationId: create_dataset_v2
      tags:
      - Datasets
      summary: Create a dataset
      description: Creates a dataset that applies a shared parse / extraction configuration across the files submitted to it. MODELED request body.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/DatasetInput'
      responses:
        '200':
          description: The created dataset.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Dataset'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '422':
          $ref: '#/components/responses/ValidationError'
    get:
      operationId: list_datasets_v2
      tags:
      - Datasets
      summary: List datasets
      description: Lists datasets in the current project.
      responses:
        '200':
          description: A page of datasets.
          content:
            application/json:
              schema:
                type: object
                properties:
                  items:
                    type: array
                    items:
                      $ref: '#/components/schemas/Dataset'
                  next_cursor:
                    type: string
        '401':
          $ref: '#/components/responses/Unauthorized'
  /documents/v2/datasets/{dataset_id}:
    parameters:
    - $ref: '#/components/parameters/DatasetId'
    get:
      operationId: get_dataset_v2
      tags:
      - Datasets
      summary: Get a dataset
      description: Retrieves a dataset and its configuration by id.
      responses:
        '200':
          description: The requested dataset.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Dataset'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '404':
          $ref: '#/components/responses/NotFound'
    put:
      operationId: update_dataset_v2
      tags:
      - Datasets
      summary: Update a dataset
      description: Updates a dataset's parse / extraction configuration. MODELED request body.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/DatasetInput'
      responses:
        '200':
          description: The updated dataset.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Dataset'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '404':
          $ref: '#/components/responses/NotFound'
    delete:
      operationId: delete_dataset_v2
      tags:
      - Datasets
      summary: Delete a dataset
      description: Deletes a dataset.
      responses:
        '200':
          description: Deletion confirmation.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/DeleteResponse'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '404':
          $ref: '#/components/responses/NotFound'
  /documents/v2/datasets/{dataset_id}/data:
    parameters:
    - $ref: '#/components/parameters/DatasetId'
    get:
      operationId: get_dataset_data_v2
      tags:
      - Datasets
      summary: Get dataset data
      description: Retrieves the accumulated parse and structured extraction output for all documents processed under a dataset.
      parameters:
      - $ref: '#/components/parameters/Cursor'
      - $ref: '#/components/parameters/Limit'
      responses:
        '200':
          description: A page of dataset rows.
          content:
            application/json:
              schema:
                type: object
                properties:
                  items:
                    type: array
                    items:
                      type: object
                      additionalProperties: true
                  next_cursor:
                    type: string
        '401':
          $ref: '#/components/responses/Unauthorized'
        '404':
          $ref: '#/components/responses/NotFound'
  /documents/v2/datasets/{dataset_id}/parse:
    parameters:
    - $ref: '#/components/parameters/DatasetId'
    post:
      operationId: parse_dataset_file_v2
      tags:
      - Datasets
      summary: Submit a file to a dataset
      description: Submits a file to a dataset, parsing / extracting it with the dataset's shared configuration. Returns a parse id. MODELED request body.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/DatasetParseRequest'
      responses:
        '200':
          description: The parse job was accepted.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ParseJobCreated'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '422':
          $ref: '#/components/responses/ValidationError'
components:
  parameters:
    Limit:
      name: limit
      in: query
      required: false
      description: Maximum number of items to return.
      schema:
        type: integer
    DatasetId:
      name: dataset_id
      in: path
      required: true
      description: The id of the dataset.
      schema:
        type: string
    Cursor:
      name: cursor
      in: query
      required: false
      description: Pagination cursor returned by a previous list call.
      schema:
        type: string
  responses:
    ValidationError:
      description: The request payload failed validation.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/Error'
    Unauthorized:
      description: Missing or invalid Bearer token.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/Error'
    NotFound:
      description: The requested resource was not found.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/Error'
  schemas:
    DatasetInput:
      type: object
      description: MODELED dataset configuration.
      properties:
        name:
          type: string
        description:
          type: string
        parsing_options:
          $ref: '#/components/schemas/ParsingOptions'
        structured_extraction_options:
          type: array
          items:
            $ref: '#/components/schemas/StructuredExtractionOptions'
    DeleteResponse:
      type: object
      properties:
        id:
          type: string
        deleted:
          type: boolean
    Dataset:
      allOf:
      - $ref: '#/components/schemas/DatasetInput'
      - type: object
        properties:
          id:
            type: string
          created_at:
            type: string
            format: date-time
    ParseJobCreated:
      type: object
      description: Response returned when an async job is accepted. MODELED.
      properties:
        parse_id:
          type: string
          description: Identifier used to retrieve the job result.
        status:
          type: string
          enum:
          - pending
          - processing
          - successful
          - failure
    DatasetParseRequest:
      type: object
      description: MODELED request to submit a file to a dataset.
      properties:
        file_id:
          type: string
        file_url:
          type: string
          format: uri
    StructuredExtractionOptions:
      type: object
      description: MODELED structured extraction configuration.
      properties:
        schema_name:
          type: string
        json_schema:
          type: object
          additionalProperties: true
          description: A JSON Schema object describing the fields to extract.
    ParsingOptions:
      type: object
      description: MODELED parsing configuration.
      properties:
        chunking_strategy:
          type: string
          enum:
          - none
          - page
          - section
          - fragment
          description: How the parsed output is split into chunks.
        table_parsing_strategy:
          type: string
          description: Strategy used to detect and parse tables.
        table_output_mode:
          type: string
          enum:
          - markdown
          - html
          - json
          description: Serialization format for detected tables.
        signature_detection:
          type: boolean
        figure_detection:
          type: boolean
    Error:
      type: object
      properties:
        error:
          type: object
          properties:
            code:
              type: string
            message:
              type: string
            details:
              type: object
              additionalProperties: true
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      bearerFormat: apiKey
      description: 'Tensorlake API key created in the Tensorlake Cloud dashboard (cloud.tensorlake.ai). Keys are prefixed tl_apiKey_ and are passed as Authorization: Bearer <token>.'