deepset-ai--haystack/docs-website/reference/integrations-api/paddleocr.md

---
title: "PaddleOCR"
id: integrations-paddleocr
description: "PaddleOCR integration for Haystack"
slug: "/integrations-paddleocr"
---


## haystack_integrations.components.converters.paddleocr.paddleocr_vl_document_converter

### PaddleOCRVLDocumentConverter

Extracts text from documents using PaddleOCR's official document parsing API.

Uses `PaddleOCRClient` to parse documents via the PaddleOCR serving API.
For more information, please refer to:
https://www.paddleocr.ai/latest/en/version3.x/algorithm/PaddleOCR-VL/PaddleOCR-VL.html

**Usage Example:**

```python
from haystack_integrations.components.converters.paddleocr import PaddleOCRVLDocumentConverter

converter = PaddleOCRVLDocumentConverter(
    base_url="http://xxxxx.aistudio-app.com",
)
result = converter.run(sources=["sample.pdf"])
documents = result["documents"]
raw_responses = result["raw_paddleocr_responses"]
```

#### __init__

```python
__init__(
    *,
    base_url: str | None = None,
    access_token: Secret = Secret.from_env_var(
        ["PADDLEOCR_ACCESS_TOKEN", "AISTUDIO_ACCESS_TOKEN"]
    ),
    model: Model | str = Model.PADDLE_OCR_VL_16,
    file_type: FileTypeInput = None,
    use_doc_orientation_classify: bool | None = False,
    use_doc_unwarping: bool | None = False,
    use_layout_detection: bool | None = None,
    use_chart_recognition: bool | None = None,
    use_seal_recognition: bool | None = None,
    use_ocr_for_image_block: bool | None = None,
    layout_threshold: float | dict | None = None,
    layout_nms: bool | None = None,
    layout_unclip_ratio: float | list | dict | None = None,
    layout_merge_bboxes_mode: str | dict | None = None,
    layout_shape_mode: str | None = None,
    prompt_label: str | None = None,
    format_block_content: bool | None = None,
    repetition_penalty: float | None = None,
    temperature: float | None = None,
    top_p: float | None = None,
    min_pixels: int | None = None,
    max_pixels: int | None = None,
    max_new_tokens: int | None = None,
    merge_layout_blocks: bool | None = None,
    markdown_ignore_labels: list[str] | None = None,
    vlm_extra_args: dict | None = None,
    prettify_markdown: bool | None = None,
    show_formula_number: bool | None = None,
    restructure_pages: bool | None = None,
    merge_tables: bool | None = None,
    relevel_titles: bool | None = None,
    visualize: bool | None = None,
    additional_params: dict[str, Any] | None = None
) -> None
```

Create a `PaddleOCRVLDocumentConverter` component.

**Parameters:**

- **base_url** (<code>str | None</code>) – Base URL for the PaddleOCR API. Falls back to `PADDLEOCR_BASE_URL`
  env var, then the SDK default.
- **access_token** (<code>Secret</code>) – PaddleOCR access token. Falls back to `PADDLEOCR_ACCESS_TOKEN` env var.
- **model** (<code>Model | str</code>) – Document parsing model. Defaults to `Model.PADDLE_OCR_VL_16`.
- **file_type** (<code>FileTypeInput</code>) – "pdf", "image", or None for auto-detection.
- **use_doc_orientation_classify** (<code>bool | None</code>) – Enable document orientation classification.
- **use_doc_unwarping** (<code>bool | None</code>) – Enable text image unwarping.
- **use_layout_detection** (<code>bool | None</code>) – Enable layout detection.
- **use_chart_recognition** (<code>bool | None</code>) – Enable chart recognition.
- **use_seal_recognition** (<code>bool | None</code>) – Enable seal recognition.
- **use_ocr_for_image_block** (<code>bool | None</code>) – Recognize text in image blocks.
- **layout_threshold** (<code>float | dict | None</code>) – Layout detection threshold.
- **layout_nms** (<code>bool | None</code>) – Perform NMS on layout detection results.
- **layout_unclip_ratio** (<code>float | list | dict | None</code>) – Layout unclip ratio.
- **layout_merge_bboxes_mode** (<code>str | dict | None</code>) – Layout merge bounding boxes mode.
- **layout_shape_mode** (<code>str | None</code>) – Layout shape mode.
- **prompt_label** (<code>str | None</code>) – Prompt type for the VLM ("ocr", "formula", "table", "chart", "seal", "spotting").
- **format_block_content** (<code>bool | None</code>) – Format block content.
- **repetition_penalty** (<code>float | None</code>) – Repetition penalty for VLM sampling.
- **temperature** (<code>float | None</code>) – Temperature for VLM sampling.
- **top_p** (<code>float | None</code>) – Top-p for VLM sampling.
- **min_pixels** (<code>int | None</code>) – Minimum pixels for VLM preprocessing.
- **max_pixels** (<code>int | None</code>) – Maximum pixels for VLM preprocessing.
- **max_new_tokens** (<code>int | None</code>) – Maximum tokens generated by the VLM.
- **merge_layout_blocks** (<code>bool | None</code>) – Merge layout detection boxes for cross-column content.
- **markdown_ignore_labels** (<code>list\[str\] | None</code>) – Layout labels to ignore in Markdown output.
- **vlm_extra_args** (<code>dict | None</code>) – Extra configuration for the VLM.
- **prettify_markdown** (<code>bool | None</code>) – Prettify output Markdown.
- **show_formula_number** (<code>bool | None</code>) – Include formula numbers in Markdown output.
- **restructure_pages** (<code>bool | None</code>) – Restructure results across multiple pages.
- **merge_tables** (<code>bool | None</code>) – Merge tables across pages.
- **relevel_titles** (<code>bool | None</code>) – Relevel titles.
- **visualize** (<code>bool | None</code>) – Return visualization results.
- **additional_params** (<code>dict\[str, Any\] | None</code>) – Extra options passed to `PaddleOCRVLOptions.extra_options`.

#### to_dict

```python
to_dict() -> dict[str, Any]
```

Serialize the component to a dictionary.

**Returns:**

- <code>dict\[str, Any\]</code> – Dictionary with serialized data.

#### from_dict

```python
from_dict(data: dict[str, Any]) -> PaddleOCRVLDocumentConverter
```

Deserialize the component from a dictionary.

**Parameters:**

- **data** (<code>dict\[str, Any\]</code>) – Dictionary to deserialize from.

**Returns:**

- <code>PaddleOCRVLDocumentConverter</code> – Deserialized component.

#### run

```python
run(
    sources: list[str | Path | ByteStream],
    meta: dict[str, Any] | list[dict[str, Any]] | None = None,
) -> dict[str, Any]
```

Convert image or PDF files to Documents.

**Parameters:**

- **sources** (<code>list\[str | Path | ByteStream\]</code>) – List of image or PDF file paths or ByteStream objects.
- **meta** (<code>dict\[str, Any\] | list\[dict\[str, Any\]\] | None</code>) – Optional metadata to attach to the Documents. A single dict is applied
  to all documents; a list must match the number of sources.

**Returns:**

- <code>dict\[str, Any\]</code> – A dictionary with:
- `documents`: List of created Documents.
- `raw_paddleocr_responses`: List of raw PaddleOCR API responses.