Downloads · 30 days
26
22% of all-time downloads
mlx-community/Marx-3B-V2-6bit
Marx-3B-V2-6bit is a text generation model from mlx-community. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
The Model mlx-community/Marx-3B-V2-6bit was converted to MLX format from acrastt/Marx-3B-V2 using mlx-lm version 0.22.3.
Downloads · 30 days
26
22% of all-time downloads
All-time downloads
119
Public
Parameters
3.4B
2.8 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors2.8 GB · 100%
How the weights are stored.
U323.4B · 100%
From the Hugging Face model README
The Model mlx-community/Marx-3B-V2-6bit was converted to MLX format from acrastt/Marx-3B-V2 using mlx-lm version 0.22.3.
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/Marx-3B-V2-6bit")
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)