Documents and OCR

TrOCR

Recognizes a single line of text, including handwriting. The official weights are English only, but the model is often fine-tuned for other languages; there are community Russian versions.

The last open version came out in Sep 2022. The family has not been updated for a long time: the model still works, but do not expect fixes or new sizes.

Developer
Microsoft, USA
First release
Oct 2021
Latest release
Sep 2022
Sizes
62M – 608M
License
Commercial use allowedMIT
Russian
Not supported
Running
On your own serverAlso runs without a GPU
Industries
Documents and accounting, Education
Deploy on your server Hugging Face GitHub

What it does

  • Recognizing handwritten lines in questionnaires and forms
  • Recognizing printed lines after text detection on the page
  • A base for fine-tuning to your own handwriting or font

Where it is used

Document managementArchivesEducation

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions
fits
1 GPUOne GPU with 16–80 GB — mid-size versions
no versions
ClusterServer with several GPUs — flagship versions
no versions

Versions

  1. TrOCR для текста на фото (STR)
  2. TrOCR small
  3. TrOCR base и large (рукописный и печатный)

How to run it

On your own serverWeights are downloaded from Hugging Face and served with vLLM (load and API) or llama.cpp (modest hardware). The model then runs in a closed network with no per-request fees.Hugging Face
How much hardware you needCalculate the VRAM for the model size, context length and number of concurrent requests.Open the hardware calculator

I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems.

Frequently asked questions

Can TrOCR be used in a commercial project?

Yes. License: MIT. It allows commercial use, but it is still worth having a lawyer review the license before launch.

What hardware does TrOCR need?

At minimum: Laptop or regular PC, up to 8 GB of VRAM — smaller versions. Some versions also run on an ordinary CPU, without a GPU. You can calculate the exact VRAM for your model size and context in the hardware calculator.

Does TrOCR support Russian?

No. The model card lists its languages and Russian is not among them.

Where can I download TrOCR and what does it cost?

The TrOCR weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.

How I deploy it for clients

  1. SelectionI pick the model size for your task and hardware and test it on your examples.
  2. DeploymentI deploy it on your server or in a closed network and provide an API.
  3. Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
  4. IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.

Similar models

Documents and OCRPP-OCR и PP-DocLayout (классический PaddleOCR)Baidu (PaddlePaddle) · ChinaCommercial use allowed

Classic lightweight PaddleOCR models: detecting and recognizing lines of text plus page layout. They run on CPUs and phones; there is a separate model for East Slavic languages, including Russian.

DetailsDocuments and OCRSurya OCRDatalab · USACommercial use with conditions

A compact OCR toolkit from the makers of Marker and Chandra: text recognition, page layout, reading order and tables. Surya OCR 2 (650M) also runs on a CPU; Russian scored 88.8% in benchmarks.

DetailsDocuments and OCRGOT-OCR 2.0StepFun · ChinaCommercial use allowed

One of the first general-purpose new-generation OCR models: text, formulas, tables, sheet music and diagrams. Small and runs on low-end hardware, but already behind newer models.

Details

Source: github.com/microsoft/unilm/tree/master/trocr. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.