Lumina-mGPT Illuminate Flexible Photorealistic Text-to-Image Generation with Multimodal Generative Pretraining paper page: We present Lumina-mGPT, a family of multimodal autoregressive models capable of various vision and language tasks, particularly excelling in generating flexible photorealistic images from text descriptions. Unlike existing autoregressive image generation approaches, Lumina-mGPT employs a pretrained decoder-only transformer as a unified framework for modeling multimodal token sequences. Our key insight is that a simple decoder-only transformer with multimodal Generative PreTraining (mGPT), utilizing the next-token prediction objective on massive interleaved text-image sequences, can learn broad and general multimodal capabilities, thereby illuminating photorealistic text-to-image generation. Building on these pretrained models, we propose Flexible Progressive Supervised Finetuning (FP-SFT) on high-quality image-text pairs to fully unlock their potential for high-aesthetic image synthesis at any resolution while maintaining their general multimodal capabilities. Furthermore, we introduce Ominiponent Supervised Finetuning (Omni-SFT), transforming Lumina-mGPT into a foundation model that seamlessly achieves omnipotent task unification. The resulting model demonstrates versatile multimodal capabilities, including visual generation tasks like flexible text-to-image generation and controllable generation, visual recognition tasks like segmentation and depth estimation, and vision-language tasks like multiturn visual question answering. Additionally, we analyze the differences and similarities between diffusion-based and autoregressive methods in a direct comparison.

DeepSeek R1 is great. How do humans think with reasoning machines like R1? CoT-Lab represents @ii_posts's latest exploration at UBAI into cognitive partnership — where human intuition and AI reasoning become co-evolving thought partners. We enable collaboration with reasoning models: An interface to guide humans through AI's reasoning flow, actively reshape thought trajectories, and collectively elevate cognitive outcomes. True intelligence emerges not from artificial or human minds — but through their symbiotic dance. We're exploring the choreography for this new cognitive ballet. Like skilled dance partners, human and machine refine each other's moves.