NVIDIA HGX H100
NVIDIA HGX H100 now available on Vultr
Unprecedented acceleration for the world’s most
demanding AI and machine learning workloads
NVIDIA HGX H100
8 x NVIDIA H100 SXM 80GB GPUs
8 x 3.84 TB NVMe
2 x 480 GB SSD
2 x Intel Platinum 8480+
112 cores / 224 threads @ 2GHZ
2048 GB Memory
15 TB Bandwidth
100 Gbps Network
Availability
Mini-Cluster 64 H100 GPUs
Next Availability December 15, 2023
Base-Cluster 248 H100 GPUs
Next Availability January 1, 2024
Pricing
- $7.50 per GPU per hour
- Includes all components and Infiniband Networking
- Minimum term: 1 month
- Maximum term: 3 months
Key Features
- 3200 Gbps InfiniBand Networking
- Non-Blocking InfiniBand Network Design
- NVIDIA HGX H100 SXM with FP8 Support
The NVIDIA HGX H100 enables innovators and enterprises alike.
One of the key benefits of the NVIDIA HGX H100 is its ability to handle large, complex models and datasets easily. This makes it ideal for use in the healthcare, finance, and manufacturing industries, where data-intensive tasks are common.
Train up to 9x Faster than A100
Reduce compute costs with over 30x higher inference throughput, and increase productivity with over 6x improvement on typical HPC workloads.