OSUM-EChat: a end-to-end empathetic dialogue model
Adjust theme and visualize JSON input
Edit images based on text prompts
Generate talking heads from audio
Edit images based on user instructions
Generate 3D edits from 2D images
Generate animated videos from images and sketches
Fast 0.6B diffusion model aligned with HyperNoise
Generate or edit images using text prompts
Generate, edit, and understand images using text prompts