Overview
Nvidia HGX H20 Enterprise 96GB is an AI server sold by Viperatech with 96GB of GPU memory.
Key Features
- 96GB of GPU memory
- 96GB system memory
- Form factor: rack-server
Ideal For
teams who want a complete configured machine with published retail pricing
Specification
Details
GPU Architecture
NVIDIA Hopper
GPU Memory
96 GB HBM3
GPU Memory Bandwidth
4.0 TB/s
INT8
FPB Tensor Core*
BF16
FP16 Tensor Core®
TF32 Tensor Core*
74 TFLOPS
FP32
44 TFLOPS
FP64
1 TFLOPS
RT Core
N/A
MIG
Up to 7 MIG
L2 Cache
60 MB
Media Engine
7 NVDEC, 7 NVJPEG
Power
400 W
Form Factor
8-way HGX
Interconnect
PCIe Gen5 x16: 128 GB/s, NVLink: 900GB/s
$8,160.00
Prices may vary. Verify on vendor site.
Quick Specs
- Specification
- Details
- GPU Architecture
- NVIDIA Hopper
- GPU Memory
- 96 GB HBM3
- GPU Memory Bandwidth
- 4.0 TB/s
- INT8
- FPB Tensor Core*
- BF16
- FP16 Tensor Core®
Tags
llm-inferencellm-trainingsmall-modelmedium-modellarge-modelhigh-vram