


Expanding Compute for AI and HPC
With dual 5th Gen Intel® Xeon® Scalable processors, the ThinkSystem SR780a V3 offers the performance needed for compute-demanding AI and HPC tasks. It supports ample DDR5 memory, NVMe drives, and PCIe Gen5 slots to speed results.
When developing your own AI platform, or providing AI resources via the cloud, the ThinkSystem SR780a V3 delivers unparalleled compute power in a dense, compact, 5U package. Model development, training, generative AI development and large HPC simulations can all benefit from the longer sustained performance of the ThinkSystem SR780a V3.

Performance Driven by Liquid-Cooling
Data centers must excel in performance while meeting energy constraints. Lenovo's innovation is evident in the ThinkSystem SR780a V3, prioritizing energy efficiency. By using Neptune™ liquid cooling instead of traditional air cooling, you boost performance while staying within energy thresholds.
Neptune™ liquid cooling improves efficiency by removing up to 95% of heat and allows the compact 5U form factor for the ThinkSystem SR780a V3.

Ultimate Acceleration
Designed for peak GPU performance, the ThinkSystem SR780a V3 harnesses maximum acceleration with 8 NVIDIA GPUs interconnected via high-speed NVLink.
Built for AI tasks like LLM, machine learning, model training, and demanding HPC workloads. NVIDIA H200 Tensor Core GPU platforms drive unmatched acceleration across all scales, fueling top-tier elastic data centers worldwide for AI and HPC applications.