Description #

PDF content parsing operator, supports visual model parsing of PDF files and structured Markdown output.

Limitations #

• Requires access to the PDF source file (public network, Volcano Engine intranet, or TOS).
• Only processes renderable page content; if the page contains complex dynamic objects or protected content, it cannot be processed.
• For documents with extremely high resolution or a large number of pages, parsing time and resource consumption will increase significantly.
• Supports PDF documents with up to 400 pages. For documents exceeding 400 pages, it is recommended to use the start_page and num_pages parameters to request parsing in multiple batches.

Key features #

• Supports PDF page rendering and visual model parsing, outputs high-fidelity Markdown, and fully restores the original structure (heading hierarchy, tables, formulas, image regions).
• Automatically identifies image regions and returns bounding box information and pre-signed image URLs.
• Supports per-page and full-book Markdown aggregation for subsequent content processing and display.
• Supports custom artifact storage location: Specify the TOS output directory via the output_tos_path parameter. The parsing artifacts will be stored in the user-specified TOS path.

Performance notes #

• Total processing time depends on the number of pages and the network environment.
• It is recommended to access the TOS source and write TOS results within the Volcano Engine intranet environment for more stable network performance.

• Billing standards

  Subitems	Billing standards description
  Billing item	Usage is calculated based on the number of pages of the PDF/image to be parsed and billed accordingly.
  Billing type	Usage-based billing, unit: CNY/page, billed hourly based on actual usage.
  Unit price	Depends on the selected PDF parsing mode.

• Billing details
  Billing formula: Total cost = unit price * usage

  Subscenario	Unit price
  PDF parsing mode: normal mode	0.02 CNY/page
  PDF parsing mode: detail mode	0.04 CNY/page

LAS provides you with the capability for an "online experience" and offers a certain amount of free trial quota. No configuration is required, and you can directly experience the data processing capabilities of the LAS operator online.

Submit #

Interface description #

Invoke PDF document parsing (Doubao) to perform PDF content parsing. Supports visual model parsing of PDF source files from the public network, Volcano Engine intranet, or TOS, and outputs Markdown or structured output.

Request parameters #

Response parameters #

Example #

Request example

curl --location "https://operator.las.ap-southeast-1.volces.com/api/v1/submit" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer $LAS_API_KEY" \
--data '{
    "operator_id": "las_pdf_parse_doubao",
    "operator_version": "v1",
    "data": {
        "url": "https://las-ai-cn-beijing-baseline.tos-cn-beijing.volces.com/operator_cards_serving/public/baseline/las_pdf_parse_doubao/v1/pdf-sample.pdf"
    }
}'

Response example

{
    "metadata": {
        "task_id": "task-20251125163544-abc123",
        "task_status": "PENDING",
        "business_code": "200",
        "error_msg": ""
    }
}

Error codes #

Poll #

Interface description #

Query the execution status and result of the PDF parsing task.

Request parameters #

Response parameters #

Example #

Request example

curl --location "https://operator.las.ap-southeast-1.volces.com/api/v1/poll" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer $LAS_API_KEY" \
--data '{
    "operator_id": "las_pdf_parse_doubao",
    "operator_version": "v1",
    "task_id": "task-20251125163544-abc123"
}'

Response example

{
    "metadata": {
        "task_id": "task-20251125163544-abc123",
        "task_status": "COMPLETED",
        "submit_time": "2026-07-21T16:04:06+08:00",
        "end_time": "2026-07-21T16:04:06+08:00",
        "business_code": "200",
        "error_msg": ""
    },
    "data": {
        "markdown": "",
        "detail": [
            {
                "page_id": 1,
                "page_md": "",
                "page_image_hw": {
                    "h": 3508,
                    "w": 2480
                },
                "text_blocks": [
                    {
                        "text": "",
                        "label": "text",
                        "box": {
                            "x0": 100,
                            "y0": 200,
                            "x1": 400,
                            "y1": 650
                        },
                        "norm_box": [
                            0.1,
                            0.2,
                            0.4,
                            0.65
                        ]
                    }
                ]
            }
        ],
      "artifacts": {
      "markdown_tos_path": "",
      "image_tos_path": "",
      "detail_tos_path": "",
      "result_preview_url": "https://las-bucket.tos-cn-beijing.volces.com/path/to/result_preview.md?X-Amz-Expires=259200&..."
    },
    "num_pages": 1
  }
    }
}

Error codes #