TW-H100
New Taipei, Taiwan
NVIDIA H100 facility with available inventory
A Taiwan facility with 3 nodes and 24 H100 80G GPUs, optional 400G networking, air cooling, and inventory ready for delivery.
24
Total GPUs
3
Nodes (8x H100 per node)
400G
GPU Fabric (optional)
2.30 and up
USD/hr/GPU (1 or 3 year term)
Compute nodes
3 nodes with 8 NVIDIA H100 80G GPUs per node.
- GPU
- 24 units · 8x H100 80G/node
- GPU memory
- 80 GB per GPU, 1.92 TB total
- CPU
- 2x AMD EPYC 9654
- Memory
- 1,024 GB DDR5 ECC RDIMM per node
- Storage
- Data: 10 TB NVMe SSD (RAID 5+) plus 10 TB HDD (RAID 6+); the customer installs the OS drive through IPMI
Network and cooling
Air cooling with optional 400G networking.
- Network (GPU Fabric)
- Not included as standard; optional BF3 400G NIC (CX7)
- Network (External)
- 100 Mb/s standard bandwidth, burstable to 1 Gb/s with 95th percentile billing
- Cooling
- Air cooling
Note
Prices are indicative hourly rates per GPU and vary by contract length, memory option, and payment structure. Some facility cities, data center names, Tier ratings, and contracting entities are available on request and will be disclosed during contracting. The formal quote sets the final terms.
Other facilities
Compare other facilities
CZ-B300
NVIDIA B300 1,024 GPU cluster
Czech Republic · 1,024 GPU
From USD 4.25, direct liquid cooling
JP-CHIBA-H100
NVIDIA H100 plan with no down payment
Chiba, Japan · 24 GPU
From approximately USD 2.20 with no down payment
SE-H200
NVIDIA H200 facility for mixed workloads
Sweden · 296 GPU
From USD 2.88, 141 GB per GPU
Need to deploy H100 capacity quickly in Taiwan?
Available inventory is limited. Share the number of GPUs and lease term you need, and we will confirm delivery timing and pricing.