LingBot-Vision self-supervised ViT backbones (Apache-2.0) converted to timm format with bit-exact fp32 parity, staged for the timm PR.
-
belfner/vit_small_patch16_lingbot.robbyant
Image Feature Extraction • 21.6M • Updated • 259 -
belfner/vit_base_patch16_lingbot.robbyant
Image Feature Extraction • 85.7M • Updated • 138 -
belfner/vit_large_patch16_lingbot.robbyant
Image Feature Extraction • 0.3B • Updated • 128 -
belfner/vit_giant_patch16_lingbot.robbyant
Image Feature Extraction • 1B • Updated • 119