NVIDIA GH200 Superchip 对比 NVIDIA RTX A5000 — GPU 比较 (Apr 2026)
NVIDIA GH200 Superchip (96GB HBM3, 989 TFLOPS FP16, Hopper) vs NVIDIA RTX A5000 (24GB GDDR6, 32.8 TFLOPS FP16, Ampere)
|
NVIDIA GH200 Superchip
96GB HBM3 · Hopper
|
NVIDIA RTX A5000
24GB GDDR6 · Ampere
|
||
|---|---|---|---|
| 规格 | |||
| 制造商 | NVIDIA | NVIDIA | |
| 架构 | Hopper | Ampere | |
| VRAM | 96 GB HBM3 | 24 GB GDDR6 | |
| 带宽 | 4,000 GB/s | 768 GB/s | |
| FP16 (Tensor) | 989.0 TFLOPS | 32.8 TFLOPS | |
| FP32 | 494.5 TFLOPS | 27.8 TFLOPS | |
| TDP | 700W | 230W | |
| 发布年份 | 2023 | 2021 | |
| 细分市场 | Data center | Professional | |
| 适用场景 | Large-scale AI training HPC | Rendering CAD mid-scale AI | |
| 云端价格 | |||
| 最便宜的按需 | — | — | |
| 最便宜的竞价 | — | — | |
| 供应商 | 0 | 0 | |
NVIDIA GH200 Superchip
NVIDIA RTX A5000