Skip to main content

MiniMax-H3_experimental

LORA

The 4 step LoRA was trained using shift 6 and the 8 step was trained using shift 12. I made this merge to use as a video refiner only so I recommend u

MiniMax-H3_experimental