Llama 3.1
Collection
12 items
•
Updated
•
12
The Model mlx-community/Llama-3.1-70B-Instruct-bf16 was converted to MLX format from meta-llama/Llama-3.1-70B-Instruct using mlx-lm version 0.19.0.
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/Llama-3.1-70B-Instruct-bf16")
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)
Base model
meta-llama/Llama-3.1-70B