Replace the dynamic int8 file with a weight-only int8 file (#2)

Update model card

Add static INT8 channelwise MobileNetV3-small model

Upload mobilenet_v3_small_Google_Tensor_G5_apply_plugin.tflite with huggingface_hub

Restore previous model card

Upload GPU-compatible LiteRT artifacts

Update README: Add model card metadata, ImageNet-1k metrics, and LiteRT usage example (#1)

Update metadata with huggingface_hub

Upload folder using huggingface_hub

initial commit