Photos supplied with quotation
Dell
Dell PowerEdge XE9680 AI Server
Flagship 6U AI training server: 8x NVIDIA SXM GPUs on an HGX baseboard with NVSwitch, built for LLM training and dense inference.
| Availability | In stock |
|---|---|
| SKU | AXL-SRV-XE9680 |
| Minimum order | 1 unit |
| Condition | New |
| Pricing | By quotation |
Overview
The PowerEdge XE9680 is the flagship Dell AI training platform: eight NVIDIA SXM GPUs on an HGX baseboard with NVSwitch interconnect, purpose-built for large language model training and high-density inference clusters.
We supply the XE9680 configured to order: GPU option (H100 or H200), processors, memory, NVMe storage and networking are specified on your quotation. Every system is burn-in tested before dispatch, serial numbers are documented, and air or sea freight is booked with full export documentation.
Specifications
| GPUs | 8x NVIDIA H100 80 GB or H200 141 GB SXM |
|---|---|
| Interconnect | NVLink with NVSwitch, up to 900 GB/s GPU-to-GPU |
| Processors | 2x Intel Xeon Scalable 4th/5th Gen |
|---|---|
| Memory | 32x DDR5 DIMM slots |
| Bays | Up to 16x E3.S NVMe |
|---|
| Form factor | 6U rack, air cooled |
|---|---|
| Power | Redundant 2800 W PSUs |
Request a quote
Quotation for Dell PowerEdge XE9680 AI Server. We reply with a formal proforma invoice, typically within 24 hours.
More in Servers
View categoryDell
Dell PowerEdge R760
2U dual-socket rack server on 4th/5th Gen Intel Xeon Scalable, built for virtualization and database workloads.
HPE
HPE ProLiant DL380 Gen11
The industry-standard 2U platform for mixed enterprise workloads, with flexible NVMe storage options.
Supermicro
Supermicro SYS-421GE-TNRT GPU Server
4U GPU platform supporting up to 8 double-width accelerators for AI training and inference.