metadata
license: apache-2.0
language:
- en
- ja
base_model:
- Rakuten/RakutenAI-2.0-8x7B-instruct
tags:
- mixtral
The Model mlx-community/RakutenAI-2.0-8x7B-instruct was converted to MLX format from Rakuten/RakutenAI-2.0-8x7B-instruct using mlx-lm version 0.21.4. When converting to the MLX format, all safetensors files from the original repository have been renamed according to Hugging Face’s standard naming convention (Conversion details: https://note.com/jniimi/n/nd1f1cc8d4fad, Japanese only).
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/RakutenAI-2.0-8x7B-instruct")
prompt = "こんにちは！"
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)