

NVIDIA GeForce RTX 3090 wins 8 of 12 benchmarks, averaging 55% 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 GeForce RTX 3090 leads by 78% (4.18 vs 7.46 images/min); the closest fight is Qwen3 4B (39% apart); VRAM decides part of this one: NVIDIA GeForce RTX 3090 runs 8 of our 12 AI workloads while the other card runs 3, models that don't fit score zero.
| Benchmark | NVIDIA GeForce RTX 2080 Ti Founders Edition | NVIDIA GeForce RTX 3090 | Difference |
|---|---|---|---|
| Qwen3 4B tok/s | 148.46 | 206.67 | -28% |
| Llama 3.1 8B tok/s | 98.34 | 144.9 | -32% |
| Qwen2.5-Coder 14B tok/s | 0 | 78.79 | n/a |
| Qwen3 32B tok/s | 0 | 37.95 | n/a |
| Llama 3.3 70B tok/s | 0 | 0 | n/a |
| Stable Diffusion XL images/min | 4.18 | 7.46 | -44% |
| Z-Image Turbo images/min | 0 | 3.825 | n/a |
| FLUX.1 dev images/min | 0 | 0 | n/a |
| FLUX.1 Kontext dev images/min | 0 | 0 | n/a |
| Qwen-Image-Edit images/min | 0 | 0 | n/a |
| LTX-Video (distilled) frames/s | 0 | 2.17 | n/a |
| Wan 2.2 5B (720p) frames/s | 0 | 0.22 | n/a |
| NVIDIA GeForce RTX 2080 Ti Founders Edition | NVIDIA GeForce RTX 3090 | |
|---|---|---|
| VRAM | 11GB | 24GB |
| Transistors | 18,600M | 28,300M |
| Die size | 754 mm² | 628.4 mm² |
| Transistor density | 24.7 M/mm² | 45 M/mm² |
| Architecture | Turing (TU102) | Ampere (GA102) |
| Memory bandwidth | 616 GB/s | 936.2 GB/s |
| Boost clock | 1,545 MHz | 1,695 MHz |
| TDP | 250 W | 350 W |
| Launch MSRP | $1,199 | $1,499 |
| Release | 2018-09-20 | 2020-09-24 |
NVIDIA GeForce RTX 2080 Ti Founders Edition full review · NVIDIA GeForce RTX 3090 full review · All AI & Machine Learning rankings