

NVIDIA RTX PRO 6000 Blackwell Server Edition wins 12 of 12 benchmarks, averaging 277.7% 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 RTX PRO 6000 Blackwell Server Edition leads by 600% (3.76 vs 26.32 images/min); the closest fight is Qwen3 4B (104% apart); VRAM decides part of this one: NVIDIA RTX PRO 6000 Blackwell Server Edition runs 12 of our 12 AI workloads while the other card runs 3, models that don't fit score zero.
| Benchmark | NVIDIA GeForce RTX 3070 Ti | NVIDIA RTX PRO 6000 Blackwell Server Edition | Difference |
|---|---|---|---|
| Qwen3 4B tok/s | 154.65 | 315.72 | -51% |
| Llama 3.1 8B tok/s | 102.13 | 233.8 | -56% |
| Qwen2.5-Coder 14B tok/s | 0 | 135.07 | n/a |
| Qwen3 32B tok/s | 0 | 63.9 | n/a |
| Llama 3.3 70B tok/s | 0 | 32.08 | n/a |
| Stable Diffusion XL images/min | 3.76 | 26.32 | -86% |
| Z-Image Turbo images/min | 0 | 14.775 | n/a |
| FLUX.1 dev images/min | 0 | 6.75 | n/a |
| FLUX.1 Kontext dev images/min | 0 | 3.021 | n/a |
| Qwen-Image-Edit images/min | 0 | 2.62 | n/a |
| LTX-Video (distilled) frames/s | 0 | 15.64 | n/a |
| Wan 2.2 5B (720p) frames/s | 0 | 0.88 | n/a |
| NVIDIA GeForce RTX 3070 Ti | NVIDIA RTX PRO 6000 Blackwell Server Edition | |
|---|---|---|
| VRAM | 8GB | 96GB |
| Architecture | Ampere (GA104) | Blackwell |
| Memory bandwidth | 608 GB/s | 1792 GB/s |
| Boost clock | 1,770 MHz | 2,610 MHz |
| TDP | 290 W | 600 W |
| Launch MSRP | $599 | $8,565 |
| Release | 2021-06-10 | 2025-03-18 |
NVIDIA GeForce RTX 3070 Ti full review · NVIDIA RTX PRO 6000 Blackwell Server Edition full review · All AI & Machine Learning rankings