Nvidia releases NVLM-1.0-D-72B frontier-class multimodal LLM with decoder-only architecture SOTA results on vision-language and text-only tasks

→ View original post on X — @_akhaliq