Desk archive
Black Forest Labs has introduced FLUX 3, a multimodal foundation model trained simultaneously on images, video, audio, and robot action data. The model can generate…
Read more
Google DeepMind introduces GenCeption, a model repurposing video generators for core computer vision tasks, achieving state-of-the-art results with minimal training data. This development suggests video…
Read more
Microsoft Research, in collaboration with several universities, has developed Mirage, a new video world model that enhances spatial consistency and significantly reduces computational demands for…
Read more