JiT-L/16

JiT-L/16, developed by LTH14, is a denoising generative model that achieves high-quality image synthesis efficiently without using a tokenizer, pre-training, or extra loss functions.

Two-hop subgraph: this entity, every entity it directly relates to, and every entity those neighbors relate to. Drag a node, scroll to zoom, click to inspect — or click any neighbor and re-center the atlas there.