NVIDIA GB200 Superchip vs NVIDIA GeForce RTX 3090 Ti — GPU-Vergleich (Apr 2026)
NVIDIA GB200 Superchip (384GB HBM3e, 4,500 TFLOPS FP16, Blackwell) vs NVIDIA GeForce RTX 3090 Ti (24GB GDDR6X, 40 TFLOPS FP16, Ampere)
|
NVIDIA GB200 Superchip
384GB HBM3e · Blackwell
|
NVIDIA GeForce RTX 3090 Ti
24GB GDDR6X · Ampere
|
||
|---|---|---|---|
| Spezifikationen | |||
| Hersteller | NVIDIA | NVIDIA | |
| Architektur | Blackwell | Ampere | |
| VRAM | 384 GB HBM3e | 24 GB GDDR6X | |
| Bandbreite | 16,000 GB/s | 1,008 GB/s | |
| FP16 (Tensor) | 4500.0 TFLOPS | 40.0 TFLOPS | |
| FP32 | 150.0 TFLOPS | 20.0 TFLOPS | |
| TDP | 2700W | 450W | |
| Erscheinungsjahr | 2024 | 2022 | |
| Segment | Data center | Consumer | |
| Am besten geeignet für | Largest-scale AI training multi-trillion parameter models | Training fine-tuning rendering | |
| Cloud-Preise | |||
| Günstigste On-Demand | — | — | |
| Günstigste Spot | — | — | |
| Anbieter | 0 | 0 | |
NVIDIA GB200 Superchip
NVIDIA GeForce RTX 3090 Ti