

NVIDIA GeForce RTX 4080 wins 6 of 12 benchmarks, averaging 75% 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 4080 leads by 170% (3.76 vs 10.16 images/min); the closest fight is Llama 3.1 8B (24% apart); VRAM decides part of this one: NVIDIA GeForce RTX 4080 runs 6 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 GeForce RTX 4080 | Difference |
|---|---|---|---|
| Qwen3 4B tok/s | 154.65 | 201.76 | -23% |
| Llama 3.1 8B tok/s | 102.13 | 127.09 | -20% |
| Qwen2.5-Coder 14B tok/s | 0 | 69.93 | 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 | 3.76 | 10.16 | -63% |
| Z-Image Turbo images/min | 0 | 2.55 | 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 | 3.23 | n/a |
| Wan 2.2 5B (720p) frames/s | 0 | 0 | n/a |
| NVIDIA GeForce RTX 3070 Ti | NVIDIA GeForce RTX 4080 | |
|---|---|---|
| VRAM | 8GB | 16GB |
| Transistors | 17,400M | 45,900M |
| Die size | 392.5 mm² | 378.6 mm² |
| Transistor density | 44.3 M/mm² | 121.2 M/mm² |
| Architecture | Ampere (GA104) | Ada Lovelace (AD103) |
| Memory bandwidth | 608 GB/s | 716.8 GB/s |
| Boost clock | 1,770 MHz | 2,505 MHz |
| TDP | 290 W | 320 W |
| Launch MSRP | $599 | $1,199 |
| Release | 2021-06-10 | 2022-11-16 |
NVIDIA GeForce RTX 3070 Ti full review · NVIDIA GeForce RTX 4080 full review · All AI & Machine Learning rankings