Search for a command to run...
MANZANO: A Simple and Scalable Unified Multimodal Model with a Hybrid Vision Tokenizer