README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
  - mlx
  - mlx-my-repo
base_model: yanolja/EEVE-Korean-Instruct-7B-v2.0-Preview
model-index:
  - name: yanolja/EEVE-Korean-Instruct-7B-v2.0-Preview
    results: []

KYUNGYONG/EEVE-Korean-Instruct-7B-v2.0-Preview-mlx-4Bit

The Model KYUNGYONG/EEVE-Korean-Instruct-7B-v2.0-Preview-mlx-4Bit was converted to MLX format from yanolja/EEVE-Korean-Instruct-7B-v2.0-Preview using mlx-lm version 0.22.3.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("KYUNGYONG/EEVE-Korean-Instruct-7B-v2.0-Preview-mlx-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)