PP-OCRv6_small_rec

PP-OCRv6_small_rec

Reads text from images in English and Chinese, the small model in the PP-OCRv6 family for PaddleOCR. For developers who want compact OCR on their own machines.

Made by PaddlePaddle on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face.

About

What it is: PP-OCRv6_small_rec is an image to text model from PaddlePaddle for the PaddleOCR library. It belongs to the PP-OCRv6 family, which the card says spans 1.5M to 34.5M parameters. The "small" in its name marks the compact end of that family, and it handles English and Chinese.

Who it suits: It suits people who need to turn pictures of text into machine readable text, and who want a small model rather than a billion-parameter one.

Industry | Who uses it | How they use it | Example
Document handling | Back-office teams | Pull text out of scans and photos | A folder of scanned pages turned into searchable text
Software | Developers | Add text recognition to an app | An app that reads text from uploaded photos in English or Chinese
Research | OCR researchers | Compare a compact model with larger vision-language models | Setting this model beside the other sizes in its family

You can skip it if you need languages beyond English and Chinese, since those are the two the source lists.

Where to use it:

• Python with the PaddleOCR library
• Windows, macOS or Linux, where the files download
• Your own scripts or apps that need to read text from images

Use case | Who | What they get back
Read a scanned page | Office staff | The text on the page as characters
Read a photo | App developers | Text found in the picture, in English or Chinese
Small model test | Researchers | A compact model to compare with larger ones

How to set it up:

• Python and pip
• The PaddleOCR library, which the model card uses to run the model

1. Install the Hugging Face hub package.
2. Download the model files to your computer.
3. Open the model card for how to run the model with PaddleOCR.

pip install -U huggingface_hub
hf download PaddlePaddle/PP-OCRv6_small_rec

How to use it: Once the files are on disk, run the model through PaddleOCR as the model card describes and give it an image that contains text. The text it reads comes back as characters you can copy.

Tool | What it does
huggingface_hub | The Hugging Face download package
hf download | Copies the model files to your computer

Example: You point it at a photo of a printed page in English or Chinese. Because it is an image to text model, the words in the picture come back as plain text.

For newcomers and experts: This is optical character recognition, which means software that reads the letters in a picture. For a first try, download the files with the two commands above, then follow the model card to run one image through PaddleOCR.

The card's headline presents PP-OCRv6 as a family from 1.5M to 34.5M parameters that aims to beat vision-language models with billions of parameters at OCR. The part of the card available here gives no layout, speed or benchmark numbers, so check the full card before you rely on any of them.

An image that holds text
PaddleOCR loads this model
The model reads the characters
Plain text comes back

Good to know: The source here is short, so these points cover only what it states.

• License: Apache 2.0
• Cost to run: no API key or paid service is named
• Data it reaches: the images you give it, plus the download from Hugging Face, and the source names no other access
• Not for: languages other than English and Chinese, which are the two the source lists

On Hugging Face, 13 November 2026

• Task: image to text
• Library: PaddleOCR
• Downloads in the last 30 days: 10,322
• Likes: 17
• License: apache-2.0
• Languages: en, zh
• Quantized versions: 1
• Fine-tunes: 4
• Tags: paddleocr, paddlepaddle, text_recognition, image-to-text

Links

• Model: https://huggingface.co/PaddlePaddle/PP-OCRv6_small_rec
• Files: https://huggingface.co/PaddlePaddle/PP-OCRv6_small_rec/tree/main

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download PaddlePaddle/PP-OCRv6_small_rec

Then follow the usage in its model card. Nothing in this kit runs by itself.