HuggingFaceH4/ultrachat_200k
Viewer • Updated • 515k • 60.4k • 735
How to use berszi/SambaLingo-Hungarian-Chat-mlx-8Bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir SambaLingo-Hungarian-Chat-mlx-8Bit berszi/SambaLingo-Hungarian-Chat-mlx-8Bit
The Model berszi/SambaLingo-Hungarian-Chat-mlx-8Bit was converted to MLX format from sambanovasystems/SambaLingo-Hungarian-Chat using mlx-lm version 0.31.2.
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("berszi/SambaLingo-Hungarian-Chat-mlx-8Bit")
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)
8-bit
Base model
sambanovasystems/SambaLingo-Hungarian-Chat