---
title: nougat-base
canonical_url: "https://www.modelscope.cn/models/cubeai/nougat-base"
md_url: "https://www.modelscope.cn/models/cubeai/nougat-base.md"
repository: cubeai/nougat-base
chinese_name: "PDF转MarkDown(nougat-base)"
last_updated: 2024-03-06
license: "Apache License 2.0"
pipeline_tag: document-understanding
tasks:
  - document-understanding
model_type:
  - vision-encoder-decoder
architectures:
  - VisionEncoderDecoderModel
parameters: 348.7M
tensor_type:
  - F32
  - I64
library_name:
  - safetensors
  - pytorch
frameworks:
  - Pytorch
downloads: 296
stars: 10
---

# nougat-base

> nougat-base - cubeai 在 ModelScope 开源的模型。PDF转MarkDown(nougat-base)

cubeai/nougat-base 是 ModelScope 魔搭社区上的 348.7M 参数document-understanding模型，采用 Apache License 2.0 许可。

- **Repository**: cubeai/nougat-base
- **License**: Apache License 2.0
- **Tasks**: document-understanding
- **Parameters**: 348.7M
- **Downloads**: 296
- **Stars**: 10
- **Last updated**: 2024-03-06

Source: https://www.modelscope.cn/models/cubeai/nougat-base

---

# PDF转MarkDown(nougat-base)

Nougat model trained on PDF-to-markdown. It was introduced in the paper [Nougat: Neural Optical Understanding for Academic Documents](https://arxiv.org/abs/2308.13418) by Blecher et al. and first released in [this repository](https://github.com/facebookresearch/nougat/tree/main). 

Nougat is a [Donut](https://hf-mirror.com/docs/transformers/model_doc/donut) model trained to transcribe scientific PDFs into an easy-to-use markdown format. The model consists of a Swin Transformer as vision encoder, and an mBART model as text decoder.

The model is trained to autoregressively predict the markdown given only the pixels of the PDF image as input.

## 使用

https://openi.pcl.ac.cn/cubeai-model-zoo/hf_facebook_nougat-base

## 模型来源

https://hf-mirror.com/facebook/nougat-base