Verily integrated Nvidia’s GPU‑accelerated libraries and Blackwell/Hopper GPUs into its Pre precision health research platform to speed genomic and multiomics analyses for researchers, including users of the NIH All of Us Researcher Workbench. Verily said the move will cut processing times for Parabricks and B200 workflows while adding NeMo and CUDA‑X frameworks for model development. Verily CEO Stephen Gillett and Nvidia executives framed the partnership as enabling faster AI model training, omics analysis and secure microservices across Pre products such as Refinery and Exchange. The integration targets nearly 20,000 registered researchers on All of Us and other Pre deployments. The development underlines continued industry investment to couple cloud‑native research platforms with accelerated compute to shorten analytic cycles for precision medicine and translational discovery.