In partnership with Microsoft, OpenAI will deploy its o1 model—or a variant of it—on the Los Alamos National Laboratory’s newly launched Venado supercomputer, powered by NVIDIA’s Grace Hopper architecture