KAT-V1-40B-mlx-4Bit / README.md
huangang's picture
Upload README.md with huggingface_hub
4fca9f8 verified
metadata
language:
  - multilingual
license: other
license_name: kwaipilot-license
license_link: LICENSE
library_name: transformers
base_model: Kwaipilot/KAT-V1-40B
tags:
  - mlx

huangang/KAT-V1-40B-mlx-4Bit

The Model huangang/KAT-V1-40B-mlx-4Bit was converted to MLX format from Kwaipilot/KAT-V1-40B using mlx-lm version 0.22.3.

Use with mlx

pip install mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("huangang/KAT-V1-40B-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)