46 lines
1.2 KiB
Markdown
46 lines
1.2 KiB
Markdown
|
---
|
||
|
|
base_model: mpasila/Llama-3.1-Swallow-JP-EN-Translator-v1-8B
|
||
|
|
tags:
|
||
|
|
- text-generation-inference
|
||
|
|
- transformers
|
||
|
|
- unsloth
|
||
|
|
- llama
|
||
|
|
- mlx
|
||
|
|
- mlx-my-repo
|
||
|
|
license:
|
||
|
|
- llama3.3
|
||
|
|
- gemma
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
- ja
|
||
|
|
datasets:
|
||
|
|
- mpasila/ParallelFiction-Ja_En-1k-16k-Gemma-3-ShareGPT-Filtered
|
||
|
|
- NilanE/ParallelFiction-Ja_En-100k
|
||
|
|
---
|
||
|
|
|
||
|
|
# moutons/Llama-3.1-Swallow-JP-EN-Translator-v1-8B-mlx-fp16
|
||
|
|
|
||
|
|
The Model [moutons/Llama-3.1-Swallow-JP-EN-Translator-v1-8B-mlx-fp16](https://huggingface.co/moutons/Llama-3.1-Swallow-JP-EN-Translator-v1-8B-mlx-fp16) was converted to MLX format from [mpasila/Llama-3.1-Swallow-JP-EN-Translator-v1-8B](https://huggingface.co/mpasila/Llama-3.1-Swallow-JP-EN-Translator-v1-8B) using mlx-lm version **0.31.2**.
|
||
|
|
|
||
|
|
## Use with mlx
|
||
|
|
|
||
|
|
```bash
|
||
|
|
pip install mlx-lm
|
||
|
|
```
|
||
|
|
|
||
|
|
```python
|
||
|
|
from mlx_lm import load, generate
|
||
|
|
|
||
|
|
model, tokenizer = load("moutons/Llama-3.1-Swallow-JP-EN-Translator-v1-8B-mlx-fp16")
|
||
|
|
|
||
|
|
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)
|
||
|
|
```
|