arcee-ai/Arcee-Maestro-7B-Preview-MLX

arcee-ai/Arcee-Maestro-7B-Preview-MLX

The Model arcee-ai/Arcee-Maestro-7B-Preview-MLX was converted to MLX format from arcee-ai/Arcee-Maestro-7B-Preview using mlx-lm version 0.21.1.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate
from huggingface_hub import snapshot_download

path = snapshot_download(
repo_id="arcee-ai/Arcee-Maestro-7B-Preview-MLX",
allow_patterns="4bit/*", # This will download everything in the 4bit folder
local_dir="Arcee-Maestro-7B-Preview-4bit" # Optional: specify where to save
)

model, tokenizer = load("Arcee-Maestro-7B-Preview-4bit")

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)