NVIDIA GH200 Superchip vs NVIDIA GeForce RTX 4060 — Comparaison GPU (Jun 2026)
NVIDIA GH200 Superchip (96GB HBM3, 989 TFLOPS FP16, Hopper) vs NVIDIA GeForce RTX 4060 (8GB GDDR6, 15.1 TFLOPS FP16, Ada Lovelace)
|
NVIDIA GH200 Superchip
96GB HBM3 · Hopper
|
NVIDIA GeForce RTX 4060
8GB GDDR6 · Ada Lovelace
|
||
|---|---|---|---|
| Spécifications | |||
| Fabricant | NVIDIA | NVIDIA | |
| Architecture | Hopper | Ada Lovelace | |
| VRAM | 96 GB HBM3 | 8 GB GDDR6 | |
| Bande passante | 4,000 GB/s | 272 GB/s | |
| FP16 (Tensor) | 989.0 TFLOPS | 15.1 TFLOPS | |
| FP32 | 494.5 TFLOPS | 7.6 TFLOPS | |
| TDP | 700W | 115W | |
| Année de sortie | 2023 | 2023 | |
| Segment | Centre de données | GPUs grand public | |
| Idéal pour | Large-scale AI training HPC | Entry-level inference gaming | |
| Tarification Cloud | |||
| Le moins cher à la demande | — | — | |
| Le moins cher en Spot | — | — | |
| Fournisseurs | 0 | 0 | |
NVIDIA GH200 Superchip
NVIDIA GeForce RTX 4060