Instructions to use mzbac/flux1.kontext.8bit.mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mzbac/flux1.kontext.8bit.mlx with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir flux1.kontext.8bit.mlx mzbac/flux1.kontext.8bit.mlx
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- LM Studio
Running the model locally
#2
by futureopen - opened
Hi. Thanks for your work on the model. Would be able to update the instructions to run the model: mzbac/flux1.kontext.8bit.mlx locally?
The instructions on the model card seem to be outdated now, given the tool(via https://github.com/mzbac/flux.swift.cli/releases) is not flux.swift.cli, but just flux
Also the cli argument --load-quantized-path doesn't seem to exist on the tool itself