37 lines
866 B
Markdown
37 lines
866 B
Markdown
|
---
|
||
|
|
license: gemma
|
||
|
|
tags:
|
||
|
|
- mlx
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
library_name: mlx
|
||
|
|
base_model: google/gemma-3-270m-it-qat
|
||
|
|
---
|
||
|
|
|
||
|
|
# mlx-community/gemma-3-270m-it-qat-bf16
|
||
|
|
|
||
|
|
This model [mlx-community/gemma-3-270m-it-qat-bf16](https://huggingface.co/mlx-community/gemma-3-270m-it-qat-bf16) was
|
||
|
|
converted to MLX format from [google/gemma-3-270m-it-qat](https://huggingface.co/google/gemma-3-270m-it-qat)
|
||
|
|
using mlx-lm version **0.26.3**.
|
||
|
|
|
||
|
|
## Use with mlx
|
||
|
|
|
||
|
|
```bash
|
||
|
|
pip install mlx-lm
|
||
|
|
```
|
||
|
|
|
||
|
|
```python
|
||
|
|
from mlx_lm import load, generate
|
||
|
|
|
||
|
|
model, tokenizer = load("mlx-community/gemma-3-270m-it-qat-bf16")
|
||
|
|
|
||
|
|
prompt = "hello"
|
||
|
|
|
||
|
|
if tokenizer.chat_template is not None:
|
||
|
|
messages = [{"role": "user", "content": prompt}]
|
||
|
|
prompt = tokenizer.apply_chat_template(
|
||
|
|
messages, add_generation_prompt=True
|
||
|
|
)
|
||
|
|
|
||
|
|
response = generate(model, tokenizer, prompt=prompt, verbose=True)
|
||
|
|
```
|