Lenovo ThinkSystem SR650a V4
Last Updated: August 25, 2026
The Lenovo ThinkSystem SR650a V4 is an ideal 2-socket 2U rack server for customers who want to maximize GPU compute power while retaining a traditional 2U rack form factor. It supports up to four double-wide GPUs or eight single-wide GPUs at the front of the chassis, paired with one or two Intel® Xeon® 6 processors (up to 86 cores, 350W TDP) featuring Intel® Advanced Matrix Extensions (Intel® AMX) acceleration for deep learning training and inference. The purpose-built SR650i V4 Inference Model pairs dual Intel® Xeon® 6530P processors with accelerator support and 512GB memory, delivering a validated, ready-to-deploy configuration for LLM and VLM inference workloads. Customers avoid re-architecting racks or cooling for high-density GPU chassis, instead scaling AI compute within familiar 2U infrastructure. Support for high-bandwidth accelerator configurations, PCIe 5.0 throughout, and optional Neptune Core liquid cooling make it well-suited to AI lifecycle, training, and inference at data center scale.
- Product Features
- Built-In AI Acceleration
- High-Density GPU Compute
- Purpose-Built Inference Configuration
- Enterprise-Ready Scalability
Related Products
