Card: correct the GPU node counts — the previous figures added the XNNPACK fallback partition to the GPU delegate partition

Card: add a measured Performance table (Pixel 8a, benchmark_model GPU+CPU), naming the runtime for each row

Add minimal usage snippets and hero image

Upload sample-depth.png with huggingface_hub

Upload sample-depth.png with huggingface_hub

Upload convert_da3.py with huggingface_hub

Upload da3_small_gpu_fp16.tflite with huggingface_hub

Upload README.md with huggingface_hub

initial commit