Developer Ports SAM2 To Apple Silicon Using MLX For 1.25x Speedup

I am working on porting SAM models and harness into Apple silicon. Already seeing 1.25x inference speed increase on mlx with the sam2.1-small model. Quantized versions soon. Repo: https://github.com/avbiswas/sam2-mlx Model: https://huggingface.co/avbiswas/sam2.1-hiera-small-mlx-fp32