---
language:
- ru
datasets:
- IlyaGusev/saiga_scored
- IlyaGusev/saiga_preferences
license: other
license_name: yandexgpt-5-lite-8b-pretrain
license_link: LICENSE
library_name: mlx
pipeline_tag: text-generation
tags:
- mlx
base_model: IlyaGusev/saiga_yandexgpt_8b
---

# alexgusevski/saiga_yandexgpt_8b-q8-mlx

This model [alexgusevski/saiga_yandexgpt_8b-q8-mlx](https://huggingface.co/alexgusevski/saiga_yandexgpt_8b-q8-mlx) was
converted to MLX format from [IlyaGusev/saiga_yandexgpt_8b](https://huggingface.co/IlyaGusev/saiga_yandexgpt_8b)
using mlx-lm version **0.30.2**.

## Use with mlx

```bash
pip install mlx-lm
```

```python
from mlx_lm import load, generate

model, tokenizer = load("alexgusevski/saiga_yandexgpt_8b-q8-mlx")

prompt = "hello"

if tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, add_generation_prompt=True, return_dict=False,
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)
```