Accurate channel state information (CSI) is essential for spectral efficiency (SE) in 6G multiple-input multiple-output orthogonal frequency-division multiplexing (MIMO-OFDM) systems, yet conventional methods incur substantial pilot overhead. This letter introduces KronFormer, a pilot-to-prediction (P2P) Transformer with factorized spatial-temporal attention aligned to the Kronecker structure of MIMO channel correlations. Unlike unfactorized Transformers that flatten spatial dimensions, KronFormer preserves the 4-D tensor structure throughout processing and decouples spatial and temporal attention, avoiding the quadratic complexity of joint spatial-temporal processing. Simulations on COST 259 channels demonstrate that Kron-Former reduces attention complexity by 20× and mean squared error (MSE) by up to 87× over unfactorized Transformers in 8×8 MIMO. KronFormer limits SE degradation to 10–13% from 50 to 200 km/h, a 3× reduction compared to the 34–39% loss of conventional Wiener filters. The architecture enables direct multi-step inference for scalable 6G channel prediction.