README.md

metadata

license: mit
base_model: moonshotai/Kimi-Dev-72B
tags:
  - code
  - swebench
  - software
  - issue-resolving
  - mlx
library_name: mlx
pipeline_tag: text-generation

mlx-community/Kimi-Dev-72B-8bit

This model mlx-community/Kimi-Dev-72B-8bit was converted to MLX format from moonshotai/Kimi-Dev-72B using mlx-lm version 0.26.0.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("mlx-community/Kimi-Dev-72B-8bit")

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)