README.md

---
tags:
- ocr
- document-processing
- dots-mocr
- multilingual
- markdown
- uv-script
- generated
configs:
- config_name: qianfan-ocr
  data_files:
  - split: train
    path: qianfan-ocr/train-*
dataset_info:
  config_name: qianfan-ocr
  features:
  - name: image
    dtype: image
  - name: b_number
    dtype: string
  - name: page_index
    dtype: int64
  - name: source_row
    dtype: int64
  - name: markdown
    dtype: string
  - name: inference_info
    dtype: string
  splits:
  - name: train
    num_bytes: 20436246
    num_examples: 50
  download_size: 20336135
  dataset_size: 20436246
---

# Document OCR using dots.mocr

This dataset contains OCR results from images in [davanstrien/moh-bench-sample](https://huggingface.co/datasets/davanstrien/moh-bench-sample) using dots.mocr, a 3B multilingual model with SOTA document parsing and SVG generation.

## Processing Details

- **Source Dataset**: [davanstrien/moh-bench-sample](https://huggingface.co/datasets/davanstrien/moh-bench-sample)
- **Model**: [rednote-hilab/dots.mocr](https://huggingface.co/rednote-hilab/dots.mocr)
- **Number of Samples**: 50
- **Processing Time**: 11.8 min
- **Processing Date**: 2026-07-08 16:53 UTC

### Configuration

- **Image Column**: `image`
- **Output Column**: `markdown`
- **Dataset Split**: `train`
- **Batch Size**: 16
- **Prompt Mode**: ocr
- **Max Model Length**: 24,000 tokens
- **Max Output Tokens**: 24,000
- **GPU Memory Utilization**: 90.0%

## Model Information

dots.mocr is a 3B multilingual document parsing model that excels at:
- 100+ Languages — Multilingual document support
- Table extraction — Structured data recognition
- Formulas — Mathematical notation preservation
- Layout-aware — Reading order and structure preservation
- Web screen parsing — Webpage layout analysis
- Scene text spotting — Text detection in natural scenes
- SVG code generation — Charts, UI layouts, scientific figures to SVG

## Dataset Structure

The dataset contains all original columns plus:
- `markdown`: The extracted text in markdown format
- `inference_info`: JSON list tracking all OCR models applied to this dataset

## Usage

```python
from datasets import load_dataset
import json

# Load the dataset
dataset = load_dataset("{output_dataset_id}", split="train")

# Access the markdown text
for example in dataset:
    print(example["markdown"])
    break

# View all OCR models applied to this dataset
inference_info = json.loads(dataset[0]["inference_info"])
for info in inference_info:
    print(f"Column: {info['column_name']} - Model: {info['model_id']}")
```

## Reproduction

This dataset was generated using the [uv-scripts/ocr](https://huggingface.co/datasets/uv-scripts/ocr) dots.mocr script:

```bash
uv run https://huggingface.co/datasets/uv-scripts/ocr/raw/main/dots-mocr.py \
    davanstrien/moh-bench-sample \
    <output-dataset> \
    --image-column image \
    --batch-size 16 \
    --prompt-mode ocr \
    --max-model-len 24000 \
    --max-tokens 24000 \
    --gpu-memory-utilization 0.9
```

Generated with [UV Scripts](https://huggingface.co/uv-scripts)