Qi Qin

A single model now rivals specialized vision-language models in understanding, while also generating and editing images, thanks to a unified discrete diffusion framework.

A 4B-parameter model, InternVL-U, outperforms 14B-parameter models in multimodal generation and editing, proving that size isn't everything.

You can now get 4x faster text-to-image generation from masked image models like Lumina-DiMOO, without sacrificing quality, by predicting feature evolution.