Show HN: mlx-chronos - benchmark MLX inference engines on Apple Silicon

- **MLX** (ML eXperiment framework) is Apple's open-source machine-learning framework for Apple Silicon (M-series chips). It uses the unified memory architecture of those chips to run models efficiently without the separate GPU memory that NVIDIA or AMD systems need. - **Inference** is the process of actually running a trained AI model (e.g., to generate text or classify an image), as opposed to training it. "Inference engines" are the software that executes this step. - **Benchmarking** measures how fast or memory-efficient different inference engines run the same model on the same hardware — important because performance varies widely depending on the engine's software optimisations. - **mlx-chronos** is a new open-source tool that lets users compare MLX-based inference with other engines (like llama.cpp or PyTorch's MPS backend) on Apple Silicon Macs, helping developers pick the fastest setup for their AI workloads.