141GB · AI Score 65.6/100 · anchored estimate vs 51 measured cards
65.6 AI Score Includes estimates
We have not run NVIDIA GH200 Grace Hopper on our bench. These figures are anchored estimates, interpolated per workload against the 51 GPUs we did measure (confidence: high (sibling silicon)). On Llama 3.1 8B (Q4_K_M) NVIDIA GH200 Grace Hopper should deliver about 273.9 tokens/sec. Stepping up to Qwen3 32B it should hold roughly 78.2 tok/s. The full Llama 3.3 70B still runs, at about 43.5 tok/s. For image generation, SDXL should run near 18.58 it/s, and FLUX.1-dev at 4.44 it/s. All 12 workloads fit in 141GB. There is no model in our suite this card has to turn down.
| Workload | Result | Telemetry | Data |
|---|---|---|---|
| Qwen3 4B | 325.5 tok/s | estimated | Est. |
| Llama 3.1 8B | 273.9 tok/s | estimated | Est. |
| Qwen2.5-Coder 14B | 151.5 tok/s | estimated | Est. |
| Qwen3 32B | 78.2 tok/s | estimated | Est. |
| Llama 3.3 70B | 43.5 tok/s | estimated | Est. |
| Workload | Result | Telemetry | Data |
|---|---|---|---|
| Stable Diffusion XL | 37.16 images/min | estimated | Est. |
| Z-Image Turbo | 23.18 images/min | estimated | Est. |
| FLUX.1 dev | 9.51 images/min | estimated | Est. |
| Workload | Result | Telemetry | Data |
|---|---|---|---|
| FLUX.1 Kontext dev | 4.39 images/min | estimated | Est. |
| Qwen-Image-Edit | 3.76 images/min | estimated | Est. |
| Workload | Result | Telemetry | Data |
|---|---|---|---|
| LTX-Video (distilled) | 17.87 frames/s | estimated | Est. |
| Wan 2.2 5B (720p) | 1.41 frames/s | estimated | Est. |
| Architecture | Hopper |
| CUDA cores | 16,896 |
| VRAM | 141GB HBM3e |
| Memory bus | 6144-bit |
| Memory bandwidth | 4900 GB/s |
| Boost clock | 1,980 MHz |
| TDP | 700 W |
| Process | TSMC 4N |
| Interface | SXM (Grace superchip) |
| Release date | 2024-02-01 |
| Launch MSRP | $40,000 |
NVIDIA GH200 Grace Hopper scores 65.6/100, #5 of 102. It ran all 12 workloads. Figures are anchored estimates, not measurements, we flag that on every row.
100% = this card, AI & Machine Learning headline metric (AI Score). #5 of 21 desktop cards in this vertical.
| GPU | Relative | % | AI Score |
|---|---|---|---|
| NVIDIA B300 | 143% | 93.8 | |
| NVIDIA B200 | 119% | 78 | |
| NVIDIA B100 | 102% | 67 | |
| NVIDIA H100 NVL | 102% | 67 | |
| NVIDIA GH200 Grace Hopper | 100% | 65.6 | |
| NVIDIA H200 | 99% | 65 | |
| NVIDIA H100 80GB HBM3 | 95% | 62.6 | |
| NVIDIA H800 80GB | 95% | 62.6 | |
| NVIDIA RTX PRO 6000 Blackwell Server Edition | 76% | 49.9 |
← All AI & Machine Learning GPU rankings
Whole-job timings, composed from our measured per-model results on this card.
| Workflow | Time | Energy | Basis |
|---|---|---|---|
| 24-frame storyboard | 80 s | n/a | estimate, 0 of 2 stages measured |
| 60-second AI short film | 2 min | n/a | estimate, 0 of 3 stages measured |
| 6-panel comic page | 2.1 min | n/a | estimate, 0 of 3 stages measured |
| Character sheet, 12 poses | 2.8 min | n/a | estimate, 0 of 2 stages measured |
| 10 short social clips | 6.3 min | n/a | estimate, 0 of 3 stages measured |
| Full codebase review | 6.6 min | n/a | estimate, 0 of 1 stage measured |
| 40-product photo shoot | 10.2 min | n/a | estimate, 0 of 2 stages measured |
| 20 long-form articles | 10.7 min | n/a | estimate, 0 of 1 stage measured |
| 100-photo restoration batch | 22.8 min | n/a | estimate, 0 of 1 stage measured |