NVIDIA GH200 144GB HBM3e
Overview
The 144 GB HBM3e configuration of the NVIDIA GH200 Grace Hopper Superchip, and the one most cloud providers rent. It keeps the 72-core Grace CPU, the Hopper H100 GPU and the 900 GB/s NVLink-C2C link of the 96 GB part, and changes only the memory system: 144 GB of HBM3e at 4.9 TB/s in place of 96 GB of HBM3 at 4 TB/s. With up to 480 GB of LPDDR5X on the CPU side that gives 624 GB of fast-access memory in one coherent address space. Compute is unchanged, because NVIDIA publishes a single performance table for both configurations, so the 50 percent extra capacity and 23 percent extra bandwidth are the whole of the difference. That matters most where the working set rather than the arithmetic is the constraint, which is why this part shows up under long-context inference and large-embedding retrieval rather than under training benchmarks.
Performance Metrics
Peak theoretical throughput by precision type
| Precision | Bits | Peak TFLOPS | Efficiency |
|---|---|---|---|
| INT8 | 8 | 1979.0 | 1.979 TFLOPS/W |
| FP8 | 8 | 1979.0 | 1.979 TFLOPS/W |
| FP16 | 16 | 990.0 | 0.990 TFLOPS/W |
| BF16 | 16 | 990.0 | 0.990 TFLOPS/W |
| TF32 | 32 | 494.0 | 0.494 TFLOPS/W |
| FP32 | 32 | 67.0 | 0.067 TFLOPS/W |
| FP64 | 64 | 34.0 | 0.034 TFLOPS/W |
Power Specifications
TDP
1.0 kW
Max Power
1.1 kW
Power Connector
PCIe 16-pin
Cooling
Liquid
Memory Specifications
Capacity
144 GB
Type
HBM3e
Bandwidth
4900 GB/s
Interface
--
Hardware & Design
Form Factor
SXM
Architecture
Hopper
Process Node
4nm
Launch Year
2024
Variant
144GB HBM3e
Market Segment
Data Center
Full Specifications
| Memory | |
|---|---|
| VRAM | 144 GB |
| Memory Type | HBM3e |
| Bandwidth | 4.9 TB/s |
| Interconnect & I/O | |
| GPU-to-GPU | NVLink-C2C |
| Interconnect Bandwidth | 900 GB/s |
| Power & Thermal | |
| TDP | 1.0 kW |
| General | |
| Form Factor | SXM |
| Architecture | Hopper |
| Launch Year | 2024 |
Documentation & Resources
Common Use Cases
The NVIDIA GH200 is optimized for high-performance computing tasks with Hopper architecture delivering 67 TFLOPS of compute power.
Where to Rent
Compare cloud providers offering on-demand GPU instances for AI training, inference, and HPC workloads.
Browse GPU Cloud ProvidersSimilar GPUs
Stay Updated on GPU Releases
Get notified when new GPUs are added or specifications are updated.
No spam, unsubscribe anytime.