---
title: PaddleOCR-VL-GGUF
canonical_url: "https://www.modelscope.cn/models/megemini/PaddleOCR-VL-GGUF"
md_url: "https://www.modelscope.cn/models/megemini/PaddleOCR-VL-GGUF.md"
repository: megemini/PaddleOCR-VL-GGUF
chinese_name: megemini/PaddleOCR-VL-GGUF
last_updated: 2026-02-22
license: "Apache License 2.0"
pipeline_tag: image-text-to-text
tasks:
  - image-text-to-text
base_model:
  - PaddlePaddle/PaddleOCR-VL
base_model_relation: repackage
library_name:
  - gguf
  - pytorch
frameworks:
  - Pytorch
downloads: 1626
stars: 2
tags:
  - gguf
---

# PaddleOCR-VL-GGUF

> PaddleOCR-VL-GGUF - megemini 在 ModelScope 开源的模型。PaddleOCR-VL-GGUF for llama.cpp

megemini/PaddleOCR-VL-GGUF 是 ModelScope 魔搭社区上的image-text-to-text模型，采用 Apache License 2.0 许可，基于 PaddlePaddle/PaddleOCR-VL 构建。

- **Repository**: megemini/PaddleOCR-VL-GGUF
- **License**: Apache License 2.0
- **Tasks**: image-text-to-text
- **Base model**: PaddlePaddle/PaddleOCR-VL
- **Tags**: gguf
- **Downloads**: 1626
- **Stars**: 2
- **Last updated**: 2026-02-22

Source: https://www.modelscope.cn/models/megemini/PaddleOCR-VL-GGUF

---

PaddleOCR-VL GGUF for llama.cpp

PR: https://github.com/ggml-org/llama.cpp/pull/18825

Test with some images:

1. A receipt

![1640](https://github.com/user-attachments/assets/0c164f7e-875d-4dd1-a4b6-7f121a98fb16)

with command:

``` shell
./build/bin/llama-cli -m /media/shun/bigdata/Models/PaddleOCR_VL_SFT/PaddleOCR-VL-GGUF.gguf \
  --mmproj /media/shun/bigdata/Models/PaddleOCR_VL_SFT/PaddleOCR-VL-GGUF-mmproj.gguf \
  --color on\
  --image /home/shun/Pictures/1640.jpeg \
  --prompt "OCR:"
```

<img width="2486" height="1275" alt="Screenshot from 2026-01-14 13-54-35" src="https://github.com/user-attachments/assets/84ab95dd-8b71-4d88-864a-c71db7c9f823" />


2. A table

![paddleocr](https://github.com/user-attachments/assets/9bb342b9-2cda-4fdd-804f-8da0c3d87146)

with command:

``` shell
./build/bin/llama-cli -m /media/shun/bigdata/Models/PaddleOCR_VL_SFT/PaddleOCR-VL-GGUF.gguf \
  --mmproj /media/shun/bigdata/Models/PaddleOCR_VL_SFT/PaddleOCR-VL-GGUF-mmproj.gguf \
  --color on\
  --image /home/shun/Pictures/paddleocr.jpg \
  --prompt "Table Recognition:"
```

<img width="2486" height="1235" alt="Screenshot from 2026-01-14 13-56-26" src="https://github.com/user-attachments/assets/d07644d4-597b-427d-8db0-d066445726b9" />

can be formatted:

<img width="716" height="635" alt="Screenshot from 2026-01-14 13-56-53" src="https://github.com/user-attachments/assets/2ad724ed-9810-4269-949b-59019459b458" />


SDK下载
```bash
#安装ModelScope
pip install modelscope
```
```python
#SDK模型下载
from modelscope import snapshot_download
model_dir = snapshot_download('megemini/PaddleOCR-VL-GGUF')
```
Git下载
```
#Git模型下载
git clone https://www.modelscope.cn/megemini/PaddleOCR-VL-GGUF.git
```

<p style="color: lightgrey;">如果您是本模型的贡献者，我们邀请您根据<a href="https://modelscope.cn/docs/ModelScope%E6%A8%A1%E5%9E%8B%E6%8E%A5%E5%85%A5%E6%B5%81%E7%A8%8B%E6%A6%82%E8%A7%88" style="color: lightgrey; text-decoration: underline;">模型贡献文档</a>，及时完善模型卡片内容。</p>