---
title: DocOwl2
canonical_url: "https://www.modelscope.cn/models/iic/DocOwl2"
md_url: "https://www.modelscope.cn/models/iic/DocOwl2.md"
repository: iic/DocOwl2
last_updated: 2025-08-27
license: apache-2.0
pipeline_tag: image-text-to-text
tasks:
  - image-text-to-text
model_type:
  - mplug_docowl
architectures:
  - mPLUGDocOwl2
parameters: 8.6B
tensor_type:
  - BF16
library_name:
  - pytorch
  - transformer
  - safetensors
frameworks:
  - Pytorch
language:
  - en
downloads: 102684
stars: 18
tags:
  - chat
---

# DocOwl2

> DocOwl2 - iic 在 ModelScope 开源的模型。不依赖OCR进行多页文档理解的多模态大模型，单页视觉token仅324。

iic/DocOwl2 是 ModelScope 魔搭社区上的 8.6B 参数image-text-to-text模型，采用 apache-2.0 许可。

- **Repository**: iic/DocOwl2
- **License**: apache-2.0
- **Tasks**: image-text-to-text
- **Parameters**: 8.6B
- **Tags**: chat
- **Downloads**: 102684
- **Stars**: 18
- **Last updated**: 2025-08-27

Source: https://www.modelscope.cn/models/iic/DocOwl2

---

# mPLUG-DocOwl2

## Introduction
mPLUG-DocOwl2 is a state-of-the-art Multimodal LLM for OCR-free Multi-page Document Understanding. 

Through a compressing module named High-resolution DocCompressor, each page is encoded with just 324 tokens.


Github: [mPLUG-DocOwl](https://github.com/X-PLUG/mPLUG-DocOwl)

## Quickstart


```python
import torch
import os
from transformers import AutoTokenizer, AutoModel
from icecream import ic
import time

class DocOwlInfer():
    def __init__(self, ckpt_path):
        self.tokenizer = AutoTokenizer.from_pretrained(ckpt_path, use_fast=False)
        self.model = AutoModel.from_pretrained(ckpt_path, trust_remote_code=True, low_cpu_mem_usage=True, torch_dtype=torch.float16, device_map='auto')
        self.model.init_processor(tokenizer=self.tokenizer, basic_image_size=504, crop_anchors='grid_12')
        
    def inference(self, images, query):
        messages = [{'role': 'USER', 'content': '<|image|>'*len(images)+query}]
        answer = self.model.chat(messages=messages, images=images, tokenizer=self.tokenizer)
        return answer


docowl = DocOwlInfer(ckpt_path='mPLUG/DocOwl2')

images = [
        './examples/docowl2_page0.png',
        './examples/docowl2_page1.png',
        './examples/docowl2_page2.png',
        './examples/docowl2_page3.png',
        './examples/docowl2_page4.png',
        './examples/docowl2_page5.png',
    ]

answer = docowl.inference(images, query='what is this paper about? provide detailed information.')

answer = docowl.inference(images, query='what is the third page about? provide detailed information.')


```