Rumodernbert Base

🚀 RuModernBERT-base

The Russian version of the modernized bidirectional encoder-only Transformer model, offering high-performance language understanding on Russian, English, and code data.

🚀 Quick Start

The Russian version of the modernized bidirectional encoder-only Transformer model, ModernBERT. RuModernBERT was pre-trained on approximately 2 trillion tokens of Russian, English, and code data with a context length of up to 8,192 tokens, using data from the internet, books, scientific sources, and social media.

💻 Usage Examples

Basic Usage

# Don't forget to update `transformers` and install `flash-attn` if your GPU supports it.
from transformers import AutoTokenizer, AutoModelForMaskedLM

# Prepare model
model_id = "deepvk/RuModernBERT-base"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForMaskedLM.from_pretrained(model_id, attn_implementation="flash_attention_2")
model = model.eval()

# Prepare input
text = "Лимончелло это настойка из [MASK]."
inputs = tokenizer(text, return_tensors="pt")
masked_index = inputs["input_ids"][0].tolist().index(tokenizer.mask_token_id)

# Make prediction
outputs = model(**inputs)

# Show prediction
predicted_token_id = outputs.logits[0, masked_index].argmax(axis=-1)
predicted_token = tokenizer.decode(predicted_token_id)
print("Predicted token:", predicted_token)
# Predicted token:  лимона

📚 Documentation

Training Details

This is the base version with 150 million parameters and the same configuration as in ModernBERT-base. The crucial difference lies in the data we used to pre-train this model.

Tokenizer

We trained a new tokenizer following the original configuration. We maintained the size of the vocabulary and added the same special tokens. The tokenizer was trained on a mixture of Russian and English from FineWeb.

Dataset

Pre-training includes three main stages: massive pre-training, context extension, and cooldown. Unlike the original model, we did not use the same data for all stages. For the second and third stages, we used cleaner data sources.

Context length

In the first stage, the model was trained with a context length of 1,024. In the second and third stages, it was extended to 8,192.

Evaluation

To evaluate the model, we measure quality on the encodechka and Russian Super Glue (RSG) benchmarks. For RSG, we perform a grid search for optimal hyperparameters and report metrics from the dev split.

For a fair comparison, we compare the RuModernBERT model only with raw encoders that were not trained on retrieval or sentence embedding tasks.

Russian Super Glue

Encodechka

📄 License

The license for this project is apache-2.0.

📚 Citation

@misc{deepvk2025rumodernbert,
    title={RuModernBERT: Modernized BERT for Russian},
    author={Spirin, Egor and Malashenko, Boris and Sokolov Andrey},
    url={https://huggingface.co/deepvk/rumodernbert-base},
    publisher={Hugging Face}
    year={2025},
}