MANZANO: A Simple and Scalable Unified Multimodal Model with a Hybrid Vision Tokenizer Paper • 2509.16197 • Published Sep 19 • 54
