Mochi1 is a state-of-the-art text to video model with high-fidelity motion and strong prompt adherence.
