common-pile/comma_v0.1_training_dataset
Viewer • Updated • 784M • 5.51k • 41
How to use simonw/comma-v0.1-2t-mlx with MLX:
# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm
# if on a CUDA device, also pip install mlx[cuda]
# Generate text with mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("simonw/comma-v0.1-2t-mlx")
prompt = "Once upon a time in"
text = generate(model, tokenizer, prompt=prompt, verbose=True)How to use simonw/comma-v0.1-2t-mlx with MLX LM:
# Install MLX LM uv tool install mlx-lm # Generate some text mlx_lm.generate --model "simonw/comma-v0.1-2t-mlx" --prompt "Once upon a time"
See common-pile/comma-v0.1-2t for the original model card.
Try this model out using uv like this, which will download a 15GB model the first time you run it:
uv run --python 3.12 \
--with mlx-lm \
mlx_lm.generate \
--model simonw/comma-v0.1-2t-mlx \
--prompt 'Facts about pelicans:'
More notes on my blog. I created the MLX port using this command:
uv run --python 3.12 \
--with mlx-lm \
python -m mlx_lm convert \
--hf-path common-pile/comma-v0.1-2t \
--mlx-path ./comma-v0.1-2t-mlx
Quantized
Base model
common-pile/comma-v0.1-2t