Boost AI performance with C845A and C240 rack servers with PCIe GPU density or NVLink-enabled configurations. Deploy the latest NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs and manage everything with Cisco Intersight.
RTX PRO Servers for AI
Achieve enterprise reliability and performance with C845A M8 and C240 M8 configured as Cisco RTX PRO Servers—supporting NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs for LLMs, generative AI, and pro visualization.
PCIe and NVLink GPU choices
Architect for your workloads. Increase GPU-to-GPU bandwidth for training and advanced graphics pipelines utilizing high-density PCIe GPUs and NVLink-enabled topologies.
Cisco Intersight streamlines IT
Experience cloud-operated lifecycle management. Automate provisioning, firmware, monitoring, and optimization from a single dashboard across your AI fleet.
Cisco MGX platform modularity
Facilitate maintenance and grow from 2 to 8 FHFL GPUs with a Cisco UCS chassis, based on NVIDIA MGX reference architecture.
Choose your form factor
4U performance workhorse
High-density PCIe and NVLink servers
Cisco UCS C845A M8—4U rack server designed for GPU-accelerated AI and HPC with room for high-bandwidth IO and storage.
- NVIDIA RTX PRO 6000 Blackwell Server Edition support
- PCIe Gen5 GPU density; NVLink GPU choices
- Broad CPU, memory, NVMe options
- From 2 to 8 FHFL GPUs
Cisco AI servers portfolio
Dense AI GPU servers—NVIDIA HGX and AMD OAM
The Cisco UCS server platforms for AI supercomputing are designed to tackle the most demanding tasks in AI and HPC.
Cisco UCS X-Series Modular System for AI with Cisco X-Fabric Technology
Discover the X580p node with X-Fabric modular server chassis, designed for AI performance—simplifying agentic ops and data center operations.