KYUNGYONG/kanana-1.5-8b-instruct-2505-mlx-4Bit

The Model KYUNGYONG/kanana-1.5-8b-instruct-2505-mlx-4Bit was converted to MLX format from kakaocorp/kanana-1.5-8b-instruct-2505 using mlx-lm version 0.22.3.

Use with mlx

pip install mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("KYUNGYONG/kanana-1.5-8b-instruct-2505-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)
Downloads last month
45
Safetensors
Model size
1.25B params
Tensor type
F16
ยท
U32
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for KYUNGYONG/kanana-1.5-8b-instruct-2505-mlx-4Bit

Quantized
(16)
this model