Directory
Hardware
Compare GPU/CPU chips, prompt-processing (prefill) speed and inference generation speed. Token rates are measured for specific models and runtimes, not theoretical hardware specifications.
16 hardware models · Showing 1–16
| Select hardware to compare | Hardware | GPU / CPU chip | Prefill (tok/s) | Inference (tok/s) | RAM / VRAM |
|---|---|---|---|---|---|
| RTX PRO 6000 Blackwell Workstation Edition | — | — | 96 GB GPU | ||
| RTX 6000 Ada Generation | — | — | 48 GB GPU | ||
| Radeon PRO W7900 | — | — | 48 GB GPU | ||
| Radeon AI PRO R9700 | — | — | 32 GB GPU | ||
| M3 Ultra integrated GPU | — | — | 512 GB unified | ||
| Apple M4 Max 40-core GPU / Apple M4 Max | — | — | 128 GB unified | ||
| Apple M4 Pro 20-core GPU / Apple M4 Pro | — | — | 64 GB unified | ||
| NVIDIA A100 80 GB / AMD EPYC 7742 | — | — | 320 GB GPU total | ||
| NVIDIA A100 40 GB / AMD EPYC 7742 | — | — | 160 GB GPU total | ||
| Tesla V100 16 GB / Intel Xeon E5-2698 v4 | — | — | 64 GB GPU total | ||
| GB300 Blackwell Ultra / NVIDIA Grace Arm Neoverse V2 | — | — | 748 GB coherent | ||
| GB10 Blackwell / Arm Cortex-X925 + Cortex-A725 | — | — | 128 GB unified | ||
| NVIDIA GB10 Blackwell / NVIDIA Grace GB10 | — | — | 128 GB unified | ||
| Arc Pro B70 | — | — | 32 GB GPU | ||
| Arc Pro B65 | — | — | 32 GB GPU | ||
| Arc Pro B60 | — | — | 24 GB GPU |