safetensors
The plain weight-file format most labs release in, and what an MLX build actually is under the name. It stores tensors and a small header and nothing executable, which is the safe part: the older pickle-based format could run code when you loaded it.
It helps to know which of the words on a listing name a container and which name a framework. GGUF is a container: one file, weights plus tokenizer plus metadata, read by llama.cpp. MLX is a framework, and an MLX model is safetensors plus a config laid out the way that framework wants them.
So GGUF versus MLX is not really a file-format comparison. It is a choice of which engine runs the model, and the file follows from it.