Text analysisDocuments and OCR

NuExtract

Models for template-based data extraction: give it a document or scan and a JSON field template, get a filled-in JSON back. NuExtract3 (4B) also converts scans to Markdown.

Developer
NuMind, France
First release
May 2024
Latest release
Apr 2026
Sizes
0.5B – 8B
License
Commercial use allowedNuExtract 1.x and 2.0 (2B, 8B) are MIT, NuExtract3 is Apache 2.0; exception: NuExtract-2.0-4B is under the Qwen research license
Russian
Not stated
Ready-made builds
GGUF
Running
Available in OllamaAlso runs without a GPU
Industries
Documents and accounting, Finance, Legal, Retail and marketplaces
Deploy on your server Hugging Face GitHub

What it does

  • Extracting company details, amounts and dates from invoices and contracts into JSON
  • Parsing receipts, waybills and forms against a set template
  • Converting scans to Markdown for search
  • Filling CRM and accounting system records from emails

Where it is used

Accounting and document managementLegal departmentsLogistics and procurementBanking and insurance

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
fits
ClusterServer with several GPUs — flagship versions
no versions

Versions

  1. NuExtract3 (4B)
  2. NuExtract 2.0 (2B, 4B, 8B, работает с картинками)
  3. NuExtract 1.5 (многоязычная)
  4. NuExtract-large
  5. NuExtract, NuExtract-tiny

How to run it

With Ollama — the fastest wayInstall Ollama on your computer or server and find the model in its library by name. Good enough to try the model on your own tasks in one evening.Find it in the Ollama library
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. Quantization compresses a model so it takes less video memory and runs on more modest hardware. Answers change slightly, so quality is checked on your own examples.

Frequently asked questions

Can NuExtract be used in a commercial project?

Yes. License: NuExtract 1.x and 2.0 (2B, 8B) are MIT, NuExtract3 is Apache 2.0; exception: NuExtract-2.0-4B is under the Qwen research license. It allows commercial use, but it is still worth having a lawyer review the license before launch.

What hardware does NuExtract 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 NuExtract support Russian?

The model card does not list languages, so Russian support cannot be promised — it has to be tested on your own examples.

Where can I download NuExtract and what does it cost?

The NuExtract 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

Text analysisGLiNERUrchade Zaratiana and Fastino AI · France / USACommercial use allowed

Finds the entities you need in text without training: just list what to look for (name, amount, date). GLiNER2 also classifies text. Multilingual versions understand Russian.

DetailsDocuments and OCRolmOCRAi2 (Allen Institute for AI) · USACommercial use allowed

A model and toolkit for converting PDFs into clean text at scale, preserving reading order, tables and formulas. Built to process millions of pages.

DetailsDocuments and OCRSmolDocling и Granite-DoclingIBM and Hugging Face · USACommercial use allowed

Tiny models for the open Docling document converter: they turn a page into markup with tables, formulas and code. Run on an ordinary laptop.

Details

Source: huggingface.co/api/models/numind/NuExtract3. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.