

GeForce RTX 5080 wins 6 of 12 benchmarks, averaging 160.4% faster.
Both cards were measured first-party on our bench, same suite, same test rig.
The gap is widest in Stable Diffusion XL, where GeForce RTX 5080 leads by 303% (2.2 vs 8.86 images/min); the closest fight is Qwen3 4B (83% apart); VRAM decides part of this one: GeForce RTX 5080 runs 6 of our 12 AI workloads while the other card runs 3, models that don't fit score zero.
| Benchmark | NVIDIA GeForce RTX 3060 Ti | GeForce RTX 5080 | Difference |
|---|---|---|---|
| Qwen3 4B tok/s | 118.21 | 215.89 | -45% |
| Llama 3.1 8B tok/s | 76.93 | 150.71 | -49% |
| Qwen2.5-Coder 14B tok/s | 0 | 81.97 | n/a |
| Qwen3 32B tok/s | 0 | 0 | n/a |
| Llama 3.3 70B tok/s | 0 | 0 | n/a |
| Stable Diffusion XL images/min | 2.2 | 8.86 | -75% |
| Z-Image Turbo images/min | 0 | 2.7 | 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 | 3.34 | n/a |
| Wan 2.2 5B (720p) frames/s | 0 | 0 | n/a |
| NVIDIA GeForce RTX 3060 Ti | GeForce RTX 5080 | |
|---|---|---|
| VRAM | 8GB | 16GB |
| Transistors | 17,400M | 45,600M |
| Die size | 392.5 mm² | 378 mm² |
| Transistor density | 44.3 M/mm² | 120.6 M/mm² |
| Architecture | Ampere (GA104) | Blackwell (GB203) |
| Memory bandwidth | 448 GB/s | 960 GB/s |
| Boost clock | 1,665 MHz | 2,617 MHz |
| TDP | 200 W | 360 W |
| Launch MSRP | $399 | $999 |
| Release | 2020-12-02 | 2025-01-30 |
NVIDIA GeForce RTX 3060 Ti full review · GeForce RTX 5080 full review · All AI & Machine Learning rankings