schroneko's picture
88614b4ebc18a90a451a5a4f9855604694052203c8986e225f5db04ebf2fcb14
dc25399 verified
|
raw
history blame
1.03 kB
---
library_name: transformers
license: other
license_link: LICENSE
base_model: Aratako/sarashina2.1-1b-sft
tags:
- axolotl
- generated_from_trainer
- mlx
language:
- ja
model-index:
- name: sarashina2.1-1b-sft
results: []
---
# mlx-community/sarashina2.1-1b-sft-4bit
The Model [mlx-community/sarashina2.1-1b-sft-4bit](https://huggingface.co/mlx-community/sarashina2.1-1b-sft-4bit) was
converted to MLX format from [Aratako/sarashina2.1-1b-sft](https://huggingface.co/Aratako/sarashina2.1-1b-sft)
using mlx-lm version **0.20.2**.
## Use with mlx
```bash
pip install mlx-lm
```
```python
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/sarashina2.1-1b-sft-4bit")
prompt="hello"
if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
messages, tokenize=False, add_generation_prompt=True
)
response = generate(model, tokenizer, prompt=prompt, verbose=True)
```