Quanta AI Labs

DanceOPD · Text-to-Image

DanceOPD

Flow-matching distillation framework that merges multiple generative capabilities (text-to-image, local and global editing) into one model by treating each source capability as a velocity field, hard-routing one frozen field per sample, and training the student via velocity MSE on student-visited states.

Open the source ↗

Similar Vision & Image releases

Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.