Lenovo logo

Lenovo ThinkSystem SR780a V3 (HGX H100)

rack
8× H100
One NVIDIA HGX H100 8-GPU liquid-cooled baseboard, with the NVLink switches water-cooled alongside the GPUs and CPUs
FP64
272
TFLOPS
FP32
536
TFLOPS
Power
—
kW Total
Memory
—
GB Total

The Hopper configuration of Lenovo's SR780a V3, ordered under base feature code C4E1 and built on a ThinkSystem NVIDIA HGX H100 80GB 700W eight-GPU liquid-cooled board. The machine's distinguishing feature is thermal: sixth generation Lenovo Neptune water cooling on four loops cools the CPUs, the GPUs and the NVLink switches, removing up to 75 percent of the server's heat in warm water and leaving fans for everything else. That is what lets an eight-accelerator system occupy 5U instead of the 8U its air-cooled equivalents need, in a chassis 447 by 219 by 942 mm weighing up to 92 kg. Hosting is two fifth-generation Intel Xeon Scalable processors at 56 cores and 350 W, and power is eight hot-swap supplies in N+N redundancy, all standard. Lenovo publishes no system power draw, no accelerator memory total and no system-level throughput; the linked HGX H100 entry carries NVIDIA's figures for the baseboard.

We will point you at suppliers who have it. Free, and no signup.

Systems built on this design

2 configurations

Built on the NVIDIA HGX H100 8-GPU baseboard, so accelerator performance matches the reference. Chassis, CPU, memory and cooling are this vendor's own.

FP64
272
TFLOPS
FP32
536
TFLOPS
TF32
3,956
TFLOPS
FP16
7,916
TFLOPS

System Details

GPU Configuration

GPU Count: 8 GPUs
Architecture: Hopper
Interconnect: One NVIDIA HGX H100 8-GPU liquid-cooled baseboard, with the NVLink switches water-cooled alongside the GPUs and CPUs

System Specifications

Form Factor: rack
Total Power: —
Total Memory: —
Memory Bandwidth: 26.8 TB/s

Precision Performance Breakdown

PrecisionSystem PerformancePer GPUEfficiency
272 TFLOPS 34 TFLOPS 34.0 TFLOPS/kW
536 TFLOPS 67 TFLOPS 67.0 TFLOPS/kW
3,956 TFLOPS 495 TFLOPS 989.0 TFLOPS/kW
7,916 TFLOPS 990 TFLOPS 1979.0 TFLOPS/kW
7,916 TFLOPS 990 TFLOPS 1979.0 TFLOPS/kW
15,832 TFLOPS 1,979 TFLOPS 3958.0 TFLOPS/kW
15,832 TOPS 1,979 TOPS 3958.0 TFLOPS/kW

Where a vendor states a basis, the figure shown is the dense one, and any figure it publishes is listed separately. Vendors commonly headline the sparse number instead: for NVIDIA's 2:4 structured sparsity that is exactly twice the dense figure, though other vendors' sparse modes do not all follow that ratio.

Powered by NVIDIA H100

This system utilizes 8 × NVIDIA H100 SXM5 80GB GPUs, each delivering exceptional performance for AI and HPC workloads.

Per GPU TDP

700W

Per GPU Memory

80 GB

Process Node

4nm

Architecture

Hopper

Documentation & Resources

Flopper Spec Sheet

Lenovo ThinkSystem SR780a V3 (HGX H100) specifications, generated from our database. Printable.

Vendor product page

Lenovo ThinkSystem SR780a V3 (HGX H100) documentation

View ↗

Lenovo ThinkSystem SR685a V3 Server Product Guide (LP1910)

Lenovo • Latest version

View Document ↗

Typical Use Cases

AI/ML Training
High-Performance Computing
Data Analytics

The Lenovo ThinkSystem SR780a V3 (HGX H100) runs 8× H100 GPUs over One NVIDIA HGX H100 8-GPU liquid-cooled baseboard, with the NVLink switches water-cooled alongside the GPUs and CPUs, delivering 15,832 TFLOPS FP8 dense.

© 2026 Flopper.io - Compare the hardware powering AI