Overview

OCR DocVQA Donut is an OCR-free document visual question answering model designed to streamline information extraction from images. Unlike traditional pipelines that require a separate OCR engine to convert images to text before processing, Donut maps visual features directly to text sequences. This architecture significantly reduces latency and eliminates the cascading errors common in OCR-dependent workflows. For developers, this means a simpler deployment stack and better performance on structured documents like invoices, receipts, and forms where spatial layout is critical for understanding. It is an ideal choice for building automated data entry tools or intelligent document processing (IDP) systems where rapid, end-to-end inference is required.

Highlights

OCR-free architecture reduces pipeline complexity and latency
Direct visual-to-text mapping eliminates OCR transcription errors
Optimized for structured documents like receipts and invoices
MIT licensed for flexible commercial and private integration

Usage

Install

# Install Hugging Face transformers
pip install transformers torch

SDK Usage

# Load model with transformers
from transformers import AutoModel, AutoTokenizer

model = AutoModel.from_pretrained("jinhybr/OCR-DocVQA-Donut")
tokenizer = AutoTokenizer.from_pretrained("jinhybr/OCR-DocVQA-Donut")

Hugging Face Download

We recommend downloading the model via the Hugging Face CLI or Hub SDK.

Guidance：Before downloading, install huggingface_hub with:

Guidance

pip install -U huggingface_hub

CLI Download

Download the full repository

huggingface-cli download jinhybr/OCR-DocVQA-Donut

Download a single file to a local folder (e.g. config.json into ./dir)

huggingface-cli download jinhybr/OCR-DocVQA-Donut config.json --local-dir ./dir

See the official docs for more CLI options

SDK Download

# 模型下载
from huggingface_hub import snapshot_download
model_dir = snapshot_download('jinhybr/OCR-DocVQA-Donut')

Git Download

Make sure git-lfs is installed first

Git Download

git lfs install
git clone https://huggingface.co/jinhybr/OCR-DocVQA-Donut

To skip LFS large-file downloads, use:

Skip LFS

GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/jinhybr/OCR-DocVQA-Donut

Model files are hosted on the Hugging Face Hub — download directly via HF CLI / SDK / Git, not through this site.

PyTorch / Transformers Usage

Install Transformers

pip install -U transformers torch

Load the model and run inference

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained('jinhybr/OCR-DocVQA-Donut')
tokenizer = AutoTokenizer.from_pretrained('jinhybr/OCR-DocVQA-Donut')

Full Documentation

来源: HuggingFace

---
license: mit
pipeline_tag: document-question-answering
tags:

donut

image-to-text

vision

widget:

text: "What is the invoice number?"

src: "https://huggingface.co/spaces/impira/docquery/resolve/2359223c1837a7587402bda0f2643382a6eefeab/invoice.png"

text: "What is the purchase amount?"

src: "https://huggingface.co/spaces/impira/docquery/resolve/2359223c1837a7587402bda0f2643382a6eefeab/contract.jpeg"
---

Donut (base-sized model, fine-tuned on DocVQA)

Donut model fine-tuned on DocVQA. It was introduced in the paper OCR-free Document Understanding Transformer by Geewok et al. and first released in this repository.

Disclaimer: The team releasing Donut did not write a model card for this model so this model card has been written by the Hugging Face team.

Model description

Donut consists of a vision encoder (Swin Transformer) and a text decoder (BART). Given an image, the encoder first encodes the image into a tensor of embeddings (of shape batch_size, seq_len, hidden_size), after which the decoder autoregressively generates text, conditioned on the encoding of the encoder.

!model image

Intended uses & limitations

This model is fine-tuned on DocVQA, a document visual question answering dataset.

We refer to the documentation which includes code examples.