---
base_model:
- ai-sage/GigaChat-20B-A3B-instruct
language:
- ru
- en
license: mit
pipeline_tag: text-generation
library_name: transformers
---

# GigaChat-20B-A3B-instruct bf16

This model is part of the GigaChat family of Russian LLMs, based on [ai-sage/GigaChat-20B-A3B-instruct](https://huggingface.co/ai-sage/GigaChat-20B-A3B-instruct). It supports a context length of 131,000 tokens.

More details are available in [this habr article](https://habr.com/en/companies/sberdevices/articles/865996/) and the original instruct model card. The model was presented in [GigaChat Family: Efficient Russian Language Modeling Through Mixture of Experts Architecture](https://huggingface.co/papers/2506.09440).

## Example Usage with Transformers

```bash
pip install --upgrade transformers torch accelerate bitsandbytes
```

```python
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig

model_name = "ai-sage/GigaChat-20B-A3B-instruct-bf16"
tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(model_name, trust_remote_code=True, device_map="auto", torch_dtype=torch.bfloat16)
model.generation_config = GenerationConfig.from_pretrained(model_name)

messages = [
    {"role": "user", "content": "Докажи теорему о неподвижной точке"}
]
input_tensor = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt")
outputs = model.generate(input_tensor.to(model.device))

result = tokenizer.decode(outputs[0][input_tensor.shape[1]:], skip_special_tokens=False)
print(result)
```