Scale as you grow with this high-performance air-cooled server engineered for a wide range of applications like AI-ML/DL training and inferencing, advanced analytics and VDI workloads.

Key Features

Powerful and scalable for GPU workloads

Scale as you grow with this high-performance air-cooled server engineered for a wide range of applications like AI-ML/DL training and inferencing, advanced analytics and VDI workloads.

Specifications

Processor	Up to 2 x 5th Generation Intel Xeon Scalable processor with up to 64 cores per processor and with optional Intel®QuickAssist Technology
Operating System	Canonical®Ubuntu®Server LTS Microsoft Windows Server®with Hyper-V RedHat®Enterprise Linux SUSE®Linux Enterprise Server VMware®ESXi®
Chipset	Intel®C741 Chipset
Memory	DIMM Speed Up to 5600 MT/s on the 5th Generation Intel®Xeon Scalable processor Memory Type RDIMM Memory Module Slots 32 DDR5 DIMM slots Maximum RAM RDIMM 8 TB
Storage	Front Bay Up to 6 x 2.5-inch NVMe, max 92.16 TB Up to 8 x 2.5-inch SAS/SATA/NVMe, max 122.88 TB
Storage Controllers	Software RAID S160 Internal Boot Boot Optimized Storage Subsystem (BOSS-N1): HWRAID 2 x M.2 NVMe SSDs USB
Security	Cryptographically signed firmware Data at Rest Encryption (SEDs with local or external key mgmt) Secure Boot Secured Component Verification (Hardware integrity check) Secure Erase Silicon Root of Trust System Lockdown (requires iDRAC9 Enterprise or Datacenter) TPM 2.0 FIPS, CC-TCG certified, TPM 2.0 China NationZ
Management	Embedded / At-the-Server iDRAC9 iDRAC Direct iDRAC RESTful API with Redfish® iDRAC Service Module Quick Sync 2 wireless module Consoles OpenManage Enterprise OpenManage Power Manager plugin OpenManage Service plugin OpenManage Update Manager plugin CloudIQ for PowerEdge plug in OpenManage Enterprise Integration for VMware vCenter OpenManage Integration for Microsoft®System Center OpenManage Integration with Windows Admin Center Mobility OpenManage Mobile Tools iDRAC RESTful API with Redfish IPMI RACADM CLI Dell System Update Enterprise Catalogs Dell Repository Manager OpenManage Integrations BMC Truesight Microsoft System Center OpenManage Integration with ServiceNow RedHat Ansible Modules Terraform Providers VMware vCenter and vRealize Operations Manager
Power Supplies	2800 W Titanium 200—240 VAC or 240 HVDC, hot swap redundant 2400 W Platinum 100—240 VAC or 240 HVDC, hot swap redundant
Ports	Network Options 2 x 1GbE LOM card (optional) 1 x OCP card 3.0 (optional) Note:The system allows either LOM card or an OCP card or both to be installed in the system. Front Ports 1 x iDRAC Direct (Micro-AB USB) port 1 x USB 2.0 1 x VGA Rear Ports 1 x Dedicated iDRAC Ethernet port 1 x USB 2.0 1 x USB 3.0 1 x Serial (optional) 1 x VGA (optional for Direct Liquid Cooling configuration) Internal Ports 1 x USB 3.0 (optional)
Slots	1 x USB 3.0 (optional) Up to 12 PCIe slots
Bezel	Optional LCD bezel or security bezel
Form Factor	2U rack server
Dimensions & Weight	Height 3.41 inches (86.8 mm) Width 18.97 inches (482 mm) Depth 37.27 inches (946.73 mm) with bezel 36.73 inches (932.89 mm) without bezel Weight: Max 60.63 lbs. (27.5 Kg)
Rack Support	Combo (Drop-in/Slide-in) rails with SRB for 4-post racks. Note:CMA is not supported in Direct Liquid Cooling configuration
Recommended Support	Choose Dell ProSupport Plus for critical systems or Dell ProSupport for premium hardware and software support for your PowerEdge solution. Consulting and deployment offerings are also available. Contact your Dell representative today for more information. Availability and terms of Dell Services vary by region.
Regulatory	Product Safety and Environmental Datasheets Dell Regulatory Compliance Home Page Dell and the Environment

Drive AI, HPC and analytics workloads with superior performance.

Key Features

Purpose-built performance

Drive AI, HPC and analytics workloads with superior performance.

Specifications

Rapidly develop, train and deploy large machine learning models with this high-performance application server made for artificial intelligence (AI), machine learning and deep learning. Dell’s PowerEdge XE9680 delivers the industry’s best AI performance.*

Key Features

Take on demanding artificial intelligence, machine learning and deep learning

Rapidly develop, train and deploy large machine learning models with this high-performance application server made for artificial intelligence (AI), machine learning and deep learning. Dell’s PowerEdge XE9680 delivers the industry’s best AI performance.*

Specifications

Unmatched performance for AI excellence in a 4U liquid cooled chassis.

Key Features

High-performance AI inferencing and model fine-tuning

Unmatched performance for AI excellence in a 4U liquid cooled chassis.

PowerEdge XE9685L rack server resources

Featuring 24 DIMM slots for up to 3TB DDR5 ECC memory and NVMe drives optimized for NVIDIA GPUDirect storage, it accelerates data processing, ensuring faster insights and superior business outcomes.

Next level performance

Featuring 24 DIMM slots for up to 3TB DDR5 ECC memory and NVMe drives optimized for NVIDIA GPUDirect storage, it accelerates data processing, ensuring faster insights and superior business outcomes.

Scalable architecture for future growth

Factory assembly and deployment integration simplify setup, reducing costs and time-to-value. Together, these features make the XE9685L a sustainable, forward-thinking choice for modern enterprises.

Specifications

Processor	Two AMD EPYC 9005 Series processors with up to 192 cores per processor Cooling options Liquid cooled CPUs, GPUs, and NVLink Switches
Operating System	Canonical Ubuntu Server LTS
Accelerators	8 x NVIDIA HGX B200 180GB 1000W SXM6 GPUs, fully interconnected with NVIDIA NVLink technology
Memory	DIMM Speed Up to 6400 MT/s Memory Type RDIMM Memory Module Slots 24 DDR5 DIMM slots Maximum RAM RDIMM 3 TB max
Storage	Front Bays Up to 8 x 2.5-inch NVMe drives max 122.88 TB
Storage Controllers	Internal Boot Boot Optimized Storage Subsystem (NVMe BOSS-N1) HWRAID 1: 2 x M.2 SSDs Internal controllers NA
Security	Cryptographically signed firmware Data at Rest Encryption (SEDs with local or external key mgmt) Secure Boot Secured Component Verification (Hardware integrity check) Secure Erase Silicon Root of Trust System Lockdown (requires iDRAC9 Enterprise or Datacenter) TPM 2.0 FIPS, CC-TCG certified
Management	Embedded / At-the-Server iDRAC9 iDRAC Direct iDRAC RESTful API with Redfish iDRAC Service Module Consoles OME APEX AIOps Observability OpenManage Enterprise OpenManage Power Manager plugin OpenManage Service plugin OpenManage Update Manager plugin Tools Dell System Update Dell Repository Manager Enterprise Catalogs iDRAC RESTful API with Redfish IPMI RACADM CLI OpenManage Integrations OpenManage Integration with ServiceNow Red Hat Ansible Modules Terraform Providers
Power Supplies	3000W Titanium 200-240 VAC or 240 VDC, hot swap redundant for B200 GPU Fans Up to six high performance (HPR) gold grade fans installed on the top layer Up to six high performance (HPR) gold grade fans installed on the bottom layer of the system
Ports	Network Options 1 x OCP 3.0 (x8 PCIe lanes) Front Ports 1 x iDRAC Direct (Micro-AB USB) port 1 x USB 2.0 1 x VGA Rear Ports 1 x USB 2.0 1 x USB 3.0 1 x iDRAC Direct ethernet port
Slots	PCIe Up to 12 x16 Gen5 (x16 PCIe) full-height, half-length Bezel Optional LCD bezel or security bezel
Form Factor	4U rack server
Dimensions & Weight	Height 6.86 in. (174.3 mm) Width Upper 2U: 17.59 in. (447 mm) Lower 2U: 17.08 in. (434 mm) Depth 40.82 in. (1037.57 mm) with bezel 40.37 in. (1025.62 mm) without bezel Weight 208.55 lbs. (Up to 94.6 kg)
Rack Support	Your system supports the Generic Tool-less stab-in sliding rail. The Generic Tool-less stab-in sliding rail supports tool-less installation in 19″ EIA-310-E compliant square hole and unthreaded round hole 4-post racks. It allows for the partial extension of the system out of the rack, enabling serviceability of key internal components.
Recommended Support	Choose Dell ProSupport Plus or Dell ProSupport for premium hardware and software support. Consulting and deployment offerings are also available. Contact your Dell representative today for more information. Availability and terms of Dell Services vary by region. For more information, visitwww.delltechnologies.com/Services.

Unmatched Performance for AI Excellence in a 4U liquid cooled chassis.

Key Features

AI training, large-scale inferencing and High Performance Computing

Unmatched Performance for AI Excellence in a 4U liquid cooled chassis.

PowerEdge XE9680L rack server resources

Additionally, Dell’s rack-scale factory integration streamlines deployment, ensuring a reliable and efficient setup process that accelerates time-to-value for your AI initiatives.

Scalable architecture for future growth

Additionally, Dell’s rack-scale factory integration streamlines deployment, ensuring a reliable and efficient setup process that accelerates time-to-value for your AI initiatives.

Specifications

Processor	Two 5th Generation Intel Xeon Scalable processors with up to 64 cores per processor Cooling options Liquid cooled CPUs, GPUs, and NVLink Switches
Operating System	Canonical Ubuntu Server LTS
Chipset	Intel®C741 chipset
Accelerators	8 x NVIDIA HGX H200 141GB 700W SXM5 GPUs, fully interconnected with NVIDIA NVLink technology** 8 x NVIDIA HGX B200 180GB 1000W SXM5 GPUs, fully interconnected with NVIDIA NVLink technology ** Note: The NVIDIA HGX H200 GPU is available post-RTS.
Memory	DIMM Speed Up to 5600 MT/s on the 5th Generation Intel® Xeon Scalable processors at 1 DPC Up to 4400 MT/s on the 5th Generation Intel® Xeon Scalable processors at 2 DPC Memory Type RDIMM Memory Module Slots 32 DDR5 DIMM slots Maximum RAM RDIMM 4 TB max
Storage	Front Bays Up to 8 x 2.5-inch NVMe drives max 122.88 TB
Storage Controllers	Internal Boot Boot Optimized Storage Subsystem (NVMe BOSS-N1): HWRAID 1, 2 x M.2 SSDs Internal controllers NA
Security	Cryptographically signed firmware Data at Rest Encryption (SEDs with local or external key mgmt) Secure Boot Secured Component Verification (Hardware integrity check) Secure Erase Silicon Root of Trust System Lockdown (requires iDRAC9 Enterprise or Datacenter) TPM 2.0 FIPS, CC-TCG certified
Management	Embedded / At-the-Server iDRAC9 iDRAC Direct iDRAC RESTful API with Redfish iDRAC Service Module Consoles CloudIQ for PowerEdge plug in OpenManage Enterprise OpenManage Power Manager plugin OpenManage Service plugin OpenManage Update Manager plugin Tools Dell System Update Dell Repository Manager Enterprise Catalogs iDRAC RESTful API with Redfish IPMI RACADM CLI OpenManage Integrations BMC Truesight OpenManage Integration with ServiceNow Terraform Providers Red Hat Ansible Modules
Power Supplies	2800W Titanium 200-240 VAC or 240 VDC, hot swap redundant 3000W Titanium 200-240 VAC or 240 VDC, hot swap redundant for B200 GPU Fans Up to six high performance (HPR) gold grade fans installed on the top layer Up to six high performance (HPR) gold grade fans installed on the bottom layer of the system
Ports	Network Options 1 x OCP 3.0 (x8 PCIe lanes) Front Ports 1 x iDRAC Direct (Micro-AB USB) port 1 x USB 2.0 1 x VGA Rear Ports 1 x USB 2.0 1 x USB 3.0 1 x iDRAC Direct ethernet port
Slots	PCIe Up to 12 x16 Gen5 (x16 PCIe) full-height, half-length Bezel Optional LCD bezel or security bezel
Dimensions & Weight	Height 174.3 mm (6.86″) Width Upper 2U: 447 mm (17.59″) Lower 2U: 434 mm (17.08″) Depth 1037.57 mm (40.82″) with bezel 1025.62 mm (40.37″) without bezel Weight Up to 210.76 lbs (95.6 kgs)
Recommended Support	Reduce complexity and enable your IT teams to drive innovation with Dell Services. Unparalleled IT support, expert deployment, skilled management, as well as AI-powered technology that enhances your IT and end-user experiences. Contact your Dell representative today for more information. Availability and terms of Dell Services vary by region. For more information, go towww.delltechnologies.com/Services Your system supports the Generic Tool-less stab-in sliding rail. The Generic Tool-less stab-in sliding rail supports tool-less installation in 19″ EIA-310-E compliant square hole and unthreaded round hole 4-post racks. It allows for the partial extension of the system out of the rack, enabling serviceability of key internal components.

PowerEdge XE7745 empowers enterprise AI insights with flexible air-cooled PCIe GPU configurations for inferencing and model fine-tuning.

Key Features

Flexible AI acceleration

PowerEdge XE7745 empowers enterprise AI insights with flexible air-cooled PCIe GPU configurations for inferencing and model fine-tuning.

Specifications

PowerEdge XE7740 empowers enterprise AI insights with flexible air-cooled PCIe GPU configurations for inferencing and model fine-tuning.

Key Features

Flexible AI acceleration

PowerEdge XE7740 empowers enterprise AI insights with flexible air-cooled PCIe GPU configurations for inferencing and model fine-tuning.

Specifications

Redefine AI performance with liquid cooling, high-density compute, and flexible power—built for the demands of your data center.

Key Features

Flexible, advanced liquid cooling servers

Redefine AI performance with liquid cooling, high-density compute, and flexible power—built for the demands of your data center.

Specifications

Processor	Two 5th Generation AMD EPYC 9005 Series processors with up to 192 cores per processor
Operating System	Canonical Ubuntu Server LTS Red Hat Enterprise Linux SUSE Linux Enterprise Server
Accelerators	8 x AMD Instinct MI355X 288GB 1400W OAM with AMD Infinity Fabric connectivity 8 x NVIDIA HGX B300 1200W SXM7 GPUs, fully interconnected with NVIDIA NVLink technology
Memory	DIMM Speed Up to 6400 MT/s Memory Type RDIMM Memory Module Slots 24 DDR5 DIMM slots Maximum RAM RDIMM 6 TB max
Storage Controllers	Internal Controllers N/A Internal Boot Embedded BOSS: HWRAID 0/1, 2 x M.2 NVMe SSDs
Security	Cryptographically signed firmware Data at Rest Encryption (SEDs with local or external key mgmt) Secure Boot Secured Component Verification (Hardware integrity check) Secure Erase Silicon Root of Trust System Lockdown Soldered down TPM on HPM board Chassis Intrusion Detection AMD Secure Memory Encryption (SME) AMD Secure Encrypted Virtualization (SEV)
Management	Embedded / At-the-Server iDRAC10 iDRAC Direct iDRAC RESTful API with Redfish RACADM CLI iDRAC Service Module Consoles OpenManage Enterprise (OME) OME Power Manager OME Services OME Update Manager OME APEX AIOps Observability Tools IPMI OpenManage Integrations Red Hat Ansible Modules Terraform Providers
Power Supplies	33 kW Power Shelf: Comprised of 6x 5500W AC PSUs Cooling Options Liquid cooled-direct-to-chip CPUs and GPUs
Ports	Network Options 1 x OCP NIC card 3.0 (x16 PCIe lanes) Front Ports 1 x USB 3.0 1 x Mini-DisplayPort 1 x USB 2.0 Type – C port 2 x RJ45 dedicated iDRAC Ethernet port Note: All front ports are located on the FIO (front I/O) board Rear Ports N/A Internal Ports 1 x USB 3.0
Slots	PCIe Up to 8 x16 Gen5 PCIe full-heigh, half length slots for E-W traffic Up to 4 x16 Gen5 PCIe full-height, half-length slots for N-S traffic
Bezel	N/A
Form Factor	IR7000 = 44 OU (Open Rack Units) 3OU compute node Note: 1 OU = 48 mm (1.88 inch) height, and 538.98 mm (21.22 inch) width.
Dimensions & Weight	Height 5.53 inches (140.5 mm) Width 21.14 inches (537 mm) Depth 41.26 inches (1047.95 mm) with fan 35.03 inches (889.65 mm) without fan Weight 209.44 pounds (95 kg)
Rack Support	Static rail for ORv3 rack
Recommended Support	Choose Dell ProSupport Plus or Dell ProSupport for premium hardware and software support. Consulting and deployment offerings are also available. Contact your Dell representative today for more information. Availability and terms of Dell Services vary by region. For more information, visitwww.delltechnologies.com/Services.

Air-cooled AI servers designed for performance, simplicity, and seamless integration.

Key Features

Built for AI model training and inferencing

Air-cooled AI servers designed for performance, simplicity, and seamless integration.

Specifications

Rack-scale server system purpose-built for the future of AI, delivering unmatched performance and scalability to drive the next generation of AI models.

Key Features

Redefining AI infrastructure

Rack-scale server system purpose-built for the future of AI, delivering unmatched performance and scalability to drive the next generation of AI models.

Specifications