云端租用NVIDIA H100 SXM — 比较7家供应商
The flagship data center GPU for AI training. Industry standard for large language model training with 80GB HBM3 memory and NVLink interconnects.
最便宜的按需
$1.57/hr
按需平均价
$2.44/hr
最便宜的竞价
$1.49/hr
供应商
7
NVIDIA H100 SXM云端价格比较 — 7家供应商
按需
| 供应商 | 价格 / GPU / 小时 | 可用性 | 备注 | |
|---|---|---|---|---|
Latitude.sh
|
$1.57/hr | Available | GPU VM, NVLink | 访问供应商 ↗ |
Vultr
|
$1.99/hr | Available | 24-month contract | 访问供应商 ↗ |
Vast.ai
|
$2.20/hr | Available | Marketplace avg | 访问供应商 ↗ |
Massed Compute
|
$2.35/hr | Available | PCIe | 访问供应商 ↗ |
Novita AI
|
$2.59/hr | Available | — | 访问供应商 ↗ |
RunPod
|
$2.99/hr | Available | Secure Cloud, SXM | 访问供应商 ↗ |
DigitalOcean
|
$3.39/hr | Available | HGX H100 | 访问供应商 ↗ |
竞价 / 可抢占
预留
| 供应商 | 价格 / GPU / 小时 | 可用性 | 备注 | |
|---|---|---|---|---|
|
DigitalOcean
|
$2.50/hr | Available | 12-month, 8-GPU | 访问供应商 ↗ |
价格最后验证时间：April 13, 2026
NVIDIA H100 SXM 技术规格
| 制造商 | NVIDIA |
|---|---|
| 架构 | Hopper |
| VRAM | 80 GB HBM3 |
| 带宽 | 3,350 GB/s |
| FP16 (Tensor) | 990.0 TFLOPS |
| FP32 | 67.0 TFLOPS |
| TDP | 700W |
| 发布年份 | 2023 |
| 细分市场 | Data center |
| 适用场景 | Large-scale AI training, distributed workloads, LLM pre-training |