AMD Instinct MI325X vs NVIDIA GB200 Superchip — GPU-Vergleich (Apr 2026)
AMD Instinct MI325X (256GB HBM3e, 1,307 TFLOPS FP16, CDNA 3) vs NVIDIA GB200 Superchip (384GB HBM3e, 4,500 TFLOPS FP16, Blackwell). Cloud pricing: AMD Instinct MI325X from $2.00/hr. Compare specs, VRAM, performance, and pricing across 1 cloud providers to find the best GPU for your AI workload.
|
AMD Instinct MI325X
256GB HBM3e · CDNA 3
|
NVIDIA GB200 Superchip
384GB HBM3e · Blackwell
|
||
|---|---|---|---|
| Spezifikationen | |||
| Hersteller | AMD | NVIDIA | |
| Architektur | CDNA 3 | Blackwell | |
| VRAM | 256 GB HBM3e | 384 GB HBM3e | |
| Bandbreite | 6,000 GB/s | 16,000 GB/s | |
| FP16 (Tensor) | 1307.0 TFLOPS | 4500.0 TFLOPS | |
| FP32 | 163.4 TFLOPS | 150.0 TFLOPS | |
| TDP | 1000W | 2700W | |
| Erscheinungsjahr | 2024 | 2024 | |
| Segment | Data center | Data center | |
| Am besten geeignet für | AI training large model inference | Largest-scale AI training multi-trillion parameter models | |
| Cloud-Preise | |||
| Günstigste On-Demand | $2.00/hr | — | |
| Günstigste Spot | — | — | |
| Anbieter | 2 | 0 | |
| Anbieterpreise (On-Demand) | |||
|
$2.00/hr | Nicht verfügbar | |