

NVIDIA H200 wins 12 of 12 benchmarks, averaging 669.2% faster.
Both cards were measured first-party on our bench, same suite, same test rig.
The gap is widest in Stable Diffusion XL, where NVIDIA H200 leads by 1589% (2.2 vs 37.16 images/min); the closest fight is Qwen3 4B (170% apart); VRAM decides part of this one: NVIDIA H200 runs 150 of our 12 AI workloads while the other card runs 3, models that don't fit score zero.
| Benchmark | NVIDIA GeForce RTX 3060 Ti | NVIDIA H200 | Difference |
|---|---|---|---|
| Qwen3 4B tok/s | 118.21 | 318.84 | -63% |
| Llama 3.1 8B tok/s | 76.93 | 268.31 | -71% |
| Qwen2.5-Coder 14B tok/s | 0 | 148.43 | n/a |
| Qwen3 32B tok/s | 0 | 76.58 | n/a |
| Llama 3.3 70B tok/s | 0 | 42.66 | n/a |
| Stable Diffusion XL images/min | 2.2 | 37.16 | -94% |
| Z-Image Turbo images/min | 0 | 23.175 | n/a |
| FLUX.1 dev images/min | 0 | 9.514 | n/a |
| FLUX.1 Kontext dev images/min | 0 | 4.393 | n/a |
| Qwen-Image-Edit images/min | 0 | 3.76 | n/a |
| LTX-Video (distilled) frames/s | 0 | 17.87 | n/a |
| Wan 2.2 5B (720p) frames/s | 0 | 1.41 | n/a |
| NVIDIA GeForce RTX 3060 Ti | NVIDIA H200 | |
|---|---|---|
| VRAM | 8GB | 141GB |
| Architecture | Ampere (GA104) | Hopper |
| Memory bandwidth | 448 GB/s | 4800 GB/s |
| Boost clock | 1,665 MHz | 1,980 MHz |
| TDP | 200 W | 700 W |
| Launch MSRP | $399 | $31,000 |
| Release | 2020-12-02 | 2024-03-18 |
NVIDIA GeForce RTX 3060 Ti full review · NVIDIA H200 full review · All AI & Machine Learning rankings