Pathumma Vision 3.0.0-Re

Pathumma Vision 3.0.0-Re is a vision-language model developed by NECTEC for Thai OCR .

The model is based on Qwen3-VL-2B-Instruct and further trained on a large-scale OCR dataset with 377K training samples. The training focuses on improving OCR capabilities, particularly for Thai and challenging real-world document and scene-text images.

This model also uses Quantization-Aware Training (QAT) to improve the model's suitability for efficient deployment while maintaining strong OCR performance.

Model Highlights

• 🧠 Based on Qwen3-VL-2B-Instruct
• 🇹🇭 Optimized for Thai OCR
• 📚 Trained on 377K OCR samples
• ⚡ Quantization-Aware Training (QAT)
• 🖼️ Vision-language image-to-text understanding
• 📄 Designed for OCR and document/image understanding
• 🚀 Intended for efficient deployment and inference

Benchmark

We evaluate our models on ThaiOCRBench, a benchmark designed to assess OCR and document understanding capabilities across Thai and challenging real-world visual content.

ThaiOCRBench Results

Training

The model was fine-tuned on 377K OCR training samples.

Training Configuration

Training Hardware

Training was performed using:

8 × NVIDIA A100 GPUs

with gradient accumulation of 4.

Intended Use

Pathumma Vision 3.0.0-Re is intended for:

• Thai OCR
• Scene text recognition
• Document text extraction
• Thai document understanding

Quickstart

Installation

pip install -U transformers
pip install torch torchvision

Using 🤗 Transformers

import torch
from transformers import Qwen3VLForConditionalGeneration, AutoProcessor

model_id = "nectec/Pathumma-llm-vision-3.0.0-re"

model = Qwen3VLForConditionalGeneration.from_pretrained(
    model_id,
    dtype=torch.bfloat16,
    device_map="auto",
)

processor = AutoProcessor.from_pretrained(model_id)

messages = [
    {
        "role": "user",
        "content": [
            {
                "type": "image",
                "image": "path/to/your/image.jpg",
            },
            {
                "type": "text",
                "text": "อ่านข้อความในภาพนี้",
            },
        ],
    }
]

inputs = processor.apply_chat_template(
    messages,
    tokenize=True,
    add_generation_prompt=True,
    return_dict=True,
    return_tensors="pt",
)

inputs = inputs.to(model.device)

generated_ids = model.generate(
    **inputs,
    max_new_tokens=512,
)

generated_ids_trimmed = [
    out_ids[len(in_ids):]
    for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]

output_text = processor.batch_decode(
    generated_ids_trimmed,
    skip_special_tokens=True,
    clean_up_tokenization_spaces=False,
)

print(output_text[0])

Contributors

This model was developed by:

• Kun Kerdthaisong
• Thanaporn Pintobtang
• Theerawat Phromchai
• Khemjira Prachumkhong
• Teepakorn Lilek
• Theerasit Issaranon
• Sarawoot Kongyoung

Acknowledgements

We thank the NECTEC team and contributors involved in the development of Pathumma and the underlying Thai-language and vision-language resources.

This model is built upon the Qwen3-VL architecture and benefits from the work of the Qwen team.

Citation

If you find Pathumma-llm-vision-3.0.0-re useful in your research, please cite:

@misc{PathummaVision3,
  author = {
    Kerdthaisong, Kun and
    Pintobtang, Thanaporn and
    Phromchai, Theerawat and
    Prachumkhong, Khemjira and
    Lilek, Teepakorn and
    Issaranon, Theerasit and
    Kongyoung, Sarawoot
  },
  title = {Pathumma Vision 3.0.0-Re},
  year = {2026},
  url = {https://huggingface.co/nectec/Pathumma-llm-vision-3.0.0-re}
}

Please also cite the original Qwen3-VL work:

@misc{qwen3technicalreport,
  title={Qwen3 Technical Report},
  author={Qwen Team},
  year={2025},
  eprint={2505.09388},
  archivePrefix={arXiv},
  primaryClass={cs.CL},
  url={https://arxiv.org/abs/2505.09388}
}