Skip to content
mlx.app

mergekit

Model-merging toolkit that treats checkpoints like Lego bricks.

When to use it

Use this when you want to combine or distill checkpoints before quantizing, rather than shrink one existing model.

Alternatives in Quantization

mlx_lm.convert

One command turns a bf16 Hugging Face repo into a quantized MLX folder.

Core ML Tools

Apple's converter and compressor for getting models onto the Neural Engine.