NVIDIA B100 对比 NVIDIA GeForce RTX 3070 Ti — GPU 比较 (Apr 2026)
NVIDIA B100 (192GB HBM3e, 1,750 TFLOPS FP16, Blackwell) vs NVIDIA GeForce RTX 3070 Ti (8GB GDDR6X, 21.7 TFLOPS FP16, Ampere)
|
NVIDIA B100
192GB HBM3e · Blackwell
|
NVIDIA GeForce RTX 3070 Ti
8GB GDDR6X · Ampere
|
||
|---|---|---|---|
| 规格 | |||
| 制造商 | NVIDIA | NVIDIA | |
| 架构 | Blackwell | Ampere | |
| VRAM | 192 GB HBM3e | 8 GB GDDR6X | |
| 带宽 | 8,000 GB/s | 608 GB/s | |
| FP16 (Tensor) | 1750.0 TFLOPS | 21.7 TFLOPS | |
| FP32 | 60.0 TFLOPS | 10.8 TFLOPS | |
| TDP | 700W | 290W | |
| 发布年份 | 2024 | 2021 | |
| 细分市场 | Data center | Consumer | |
| 适用场景 | AI training large-scale inference | Budget inference gaming | |
| 云端价格 | |||
| 最便宜的按需 | — | — | |
| 最便宜的竞价 | — | — | |
| 供应商 | 0 | 0 | |
NVIDIA B100
NVIDIA GeForce RTX 3070 Ti