Brand: Lenovo | Category: GPUs
SKU: 4X97A94072 | Part #: 4X97A94072 | MPN: 4X97A94072
Contact for Pricing — Request a Quote
PCIe Gen5 interface and redundant hot-swap power supplies determine the server chassis and PSU compatibility requirements for this enterprise-grade accelerated computing platform. The Lenovo ThinkSystem SR670 V3 with NVIDIA H100 NVL 94GB PCIe GPU (part number 4X97A94072) is a 2U rack server engineered for AI, machine learning, and high-performance computing workloads that demand exceptional GPU density and memory bandwidth. Built on the NVIDIA Hopper architecture, this system pairs the 94GB HBM3 memory of the H100 NVL accelerator with Lenovo's proven server architecture to deliver the compute power required by data scientists, AI infrastructure teams, and hyperscale operators.
The system supports up to 8 PCIe GPU slots (configuration dependent), allowing organizations to scale their AI infrastructure horizontally while maintaining efficient power and cooling management. With 2 processor sockets for 4th Generation Intel Xeon Scalable Processors and up to 4TB of DDR5 system memory across 32 DIMM slots, this Lenovo platform delivers the CPU and memory bandwidth needed to feed multiple GPUs simultaneously. Storage flexibility accommodates up to 12 3.5-inch or 24 2.5-inch drive bays (configuration dependent) for training datasets and model checkpoints. The H100 NVL supports eight precision formats—FP8, FP16, BF16, TF32, and FP64—enabling optimized inference and training across diverse workloads. Integrated Lenovo XClarity Controller (XCC2) and XClarity Administrator provide comprehensive systems management, while UEFI Secure Boot and TPM 2.0 ensure enterprise security. Compatible with Red Hat Enterprise Linux, SUSE Linux Enterprise Server, VMware vSphere, and Ubuntu Server, this configuration meets diverse operational requirements. Contact Omnixon Global to request a quotation for the Lenovo ThinkSystem SR670 V3 with NVIDIA H100 NVL 94GB PCIe GPU (4X97A94072).
| Brand | Lenovo |
| Category | GPUs |
| SKU | 4X97A94072 |
| Part Number | 4X97A94072 |
| Condition | New |
| Manufacturer Part Number | 4X97A94072 |
| Product Line | ThinkSystem SR670 V3 |
| Form Factor | 2U Rack Server |
| GPU Model | NVIDIA H100 NVL 94GB PCIe |
| GPU Architecture | NVIDIA Hopper |
| GPU Memory | 94GB HBM3 |
| GPU Interface | PCIe Gen5 |
| Supported Precision Formats | FP8, FP16, BF16, TF32, FP64 |
| Processor Support | 4th Generation Intel Xeon Scalable Processors |
| Processor Sockets | 2 |
| Maximum System Memory | Up to 4TB DDR5 |
| Memory Slots | 32 x DIMM slots |
| GPU Slots | Up to 8 x PCIe GPU support (configuration dependent) |
| Storage Bays | Up to 12 x 3.5-inch or 24 x 2.5-inch drive bays (configuration dependent) |
| Network | Integrated 1GbE management port; PCIe expansion for high-speed networking adapters |
| Power Supply | Redundant hot-swap power supplies |
| Systems Management | Lenovo XClarity Controller (XCC2), Lenovo XClarity Administrator |
| Operating System Support | Red Hat Enterprise Linux, SUSE Linux Enterprise Server, VMware vSphere, Ubuntu Server |
| Security Features | UEFI Secure Boot, Trusted Platform Module (TPM 2.0) |
| Chassis | Standard 19-inch rack-mount, 2U |
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.
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.
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.
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.