Text analysisSearch and RAG

ruBERT, ruRoBERTa, ruELECTRA (ai-forever)

Sber's Russian-language encoders trained on large Russian corpora. A base for classifiers, NER and semantic search in Russian.

The last open version came out in Jul 2024. The family has not been updated for a long time: the model still works, but do not expect fixes or new sizes.

Developer
SberDevices (ai-forever), Russia
First release
Nov 2020
Latest release
Jul 2024
Sizes
about 30M to 430M
License
Commercial use allowedApache 2.0 and MIT; the license for ruRoBERTa-large is not stated on the model card, needs checking
Russian
Supported
Running
On your own serverAlso runs without a GPU
Industries
Software development, Documents and accounting, Customer support
Deploy on your server Hugging Face

What it does

  • Classifying requests in Russian
  • Extracting names, amounts and dates after fine-tuning
  • Detecting review sentiment

Where it is used

Banking and insuranceCustomer supportDocument management

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. ru-en-RoSBERTa
  2. ruELECTRA large
  3. ruELECTRA small и medium
  4. ruBERT base и large, ruRoBERTa-large
  5. sbert_large_nlu_ru

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 ruBERT, ruRoBERTa, ruELECTRA (ai-forever) be used in a commercial project?

Yes. License: Apache 2.0 and MIT; the license for ruRoBERTa-large is not stated on the model card, needs checking. It allows commercial use, but it is still worth having a lawyer review the license before launch.

What hardware does ruBERT, ruRoBERTa, ruELECTRA (ai-forever) 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 ruBERT, ruRoBERTa, ruELECTRA (ai-forever) support Russian?

Yes, Russian is listed on the model card.

Where can I download ruBERT, ruRoBERTa, ruELECTRA (ai-forever) and what does it cost?

The ruBERT, ruRoBERTa, ruELECTRA (ai-forever) 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 analysisrubert-tinyDavid Dale (cointegrated) · RussiaCommercial use allowed

A very small Russian-English BERT that runs fast on a regular CPU. Ready-made fine-tuned versions exist for sentiment, toxicity and emotions.

DetailsText analysisRuModernBERT и USER (deepvk)deepvk (VK) · RussiaCommercial use allowed

Russian encoders from the VK team: RuModernBERT reads long texts, USER produces vectors for search, GeRaCl classifies texts by topic without training.

DetailsSearch and RAGFRIDA / Giga-EmbeddingsSber (SberDevices) · RussiaCommercial use allowed

Sber embeddings built for Russian: according to the developers, among the best on Russian-language search benchmarks. FRIDA is compact, Giga-Embeddings is more powerful.

Details

Source: huggingface.co/ai-forever/ru-en-RoSBERTa. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.