Module kernels 

GPU kernels used by the YOLO26 model (attention, detection decode, loss).

Modules

attention

Attention kernels (flash attention, position-sensitive attention).

detect_decode

Kernel that decodes raw detection head output into boxes/scores/classes. Fused dist2bbox decode Triton kernel for YOLO detection post-processing.

loss

Loss-related kernels (CIoU, classification loss).