Lenovo ThinkSystem SR650 V3 with NVIDIA A16 64GB PCIe GPU (VDI)

Lenovo ThinkSystem SR650 V3 with NVIDIA A16 64GB PCIe GPU (VDI)

Brand: Lenovo | Category: GPUs

SKU: 4X97A68626 | Part #: 4X97A68626 | MPN: 4X97A68626

Contact for Pricing — Request a Quote

Request a Quote Contact Us

About the Lenovo ThinkSystem SR650 V3 with NVIDIA A16 64GB PCIe GPU (VDI)

Virtual Desktop Infrastructure (VDI) deployments demand GPU acceleration that balances density, memory capacity, and power efficiency—and the Lenovo ThinkSystem SR650 V3 with NVIDIA A16 64GB PCIe GPU delivers exactly that through the NVIDIA Ampere architecture (GA107 ×4 per card). The NVIDIA A16 GPU card features 64GB of GDDR6 memory (4× 16GB per physical GPU die) with up to 200 GB/s aggregate memory bandwidth, enabling high-performance virtual desktop sessions across enterprise environments. This full-height, full-length dual-slot PCIe card operates at 250W TDP and connects via PCIe Gen 4 x16, integrating seamlessly into Lenovo's 2U rack-mount SR650 V3 platform. The server itself supports dual-socket Intel Xeon Scalable 4th Generation (Sapphire Rapids) processors, up to 4TB DDR5 system memory across 32 DIMM slots, and PCIe Gen 5 connectivity, allowing configurations of up to 2× NVIDIA A16 GPUs per system. Part number 4X97A68626 ships with full vGPU support through NVIDIA vPC, NVIDIA vApps, and NVIDIA RTX vWS, enabling up to 64 vGPU instances per card using 1GB vGPU profiles for maximum user density.

IT procurement teams and infrastructure architects standardizing on virtualized desktop delivery choose this Lenovo configuration because it combines proven server architecture with purpose-built GPU acceleration for VDI workloads. The SR650 V3 platform supports leading hypervisors—VMware vSphere, Citrix Hypervisor, and Red Hat Enterprise Virtualization—ensuring compatibility with existing data center orchestration tooling and eliminating costly platform migrations.

To specify the Lenovo ThinkSystem SR650 V3 with NVIDIA A16 64GB PCIe GPU (part number 4X97A68626) for your organization, or to discuss custom configurations and volume pricing, contact Omnixon Global via RFQ today.

Technical Specifications

BrandLenovo
CategoryGPUs
SKU4X97A68626
Part Number4X97A68626
ConditionNew
Manufacturer Part Number4X97A68626
Product NameThinkSystem SR650 V3 with NVIDIA A16 64GB PCIe GPU (VDI)
GPU ModelNVIDIA A16
GPU ArchitectureNVIDIA Ampere (GA107 ×4 per card)
Total GPU Memory64GB GDDR6 (4× 16GB per physical GPU die)
GPU Memory BandwidthUp to 200 GB/s aggregate
PCIe InterfacePCIe Gen 4 x16
TDP (Thermal Design Power)250W
vGPU SupportNVIDIA vPC, NVIDIA vApps, NVIDIA RTX vWS
Max vGPU Instances per CardUp to 64 (using 1GB vGPU profiles)
Form FactorFull-height, full-length (FHFL) dual-slot PCIe
Server PlatformLenovo ThinkSystem SR650 V3
Server Form Factor2U Rack
Supported ProcessorsIntel Xeon Scalable 4th Generation (Sapphire Rapids), dual-socket
Maximum System MemoryUp to 4TB DDR5 (32× DIMM slots)
PCIe Generation (Server)PCIe Gen 5
Max GPUs per Server2× NVIDIA A16 (configuration dependent)
Supported HypervisorsVMware vSphere, Citrix Hypervisor, Red Hat Enterprise Virtualization
Primary Workload ClassificationVDI (Virtual Desktop Infrastructure) — Legacy profile support

Frequently Asked Questions about Lenovo ThinkSystem SR650 V3 with NVIDIA A16 64GB PCIe GPU (VDI)

What server platforms accept the Lenovo ThinkSystem SR650 V3 with NVIDIA A16 64GB PCIe GPU (VDI, Legacy)?

Reference servers include Dell PowerEdge XE9680 / XE9712, HPE Cray XD670, Lenovo ThinkSystem SR685a / SR675 V3, Supermicro AS-A21GE / SYS-821GE, Gigabyte G593 / G894, ASUS ESC. Share your target platform in the RFQ and we will confirm chassis-to-GPU compatibility and recommended NIC pairing.

How long is the lead time on AI GPUs?

Highly model-dependent. L40S / RTX-class: typically 3-6 weeks. H100/H200/B200 in SXM form factor: 12-16 weeks for whole-platform allocations. We quote genuine-channel ETAs only — no grey-market promises.

Do you supply matched networking (Quantum InfiniBand / Spectrum-X)?

Yes — Omnixon stocks the full NVIDIA networking lineup (Quantum-2 / Quantum-X InfiniBand, Spectrum-X Ethernet, ConnectX NICs, BlueField DPUs) so we can quote a complete training-cluster BOM, not just the GPUs.

Can you help with NVIDIA AI Enterprise licensing?

Yes. We hold genuine-channels for NVIDIA AI Enterprise software subscriptions. Add it to your RFQ and we quote node-aligned licensing along with the hardware.