

NVIDIA GeForce RTX 4090 wins 8 of 12 benchmarks, averaging 217% faster.
Both cards were measured first-party on our bench, same suite, same test rig.
The gap is widest in Z-Image Turbo, where NVIDIA GeForce RTX 4090 leads by 890% (0.75 vs 7.425 images/min); the closest fight is Qwen3 4B (32% apart); VRAM decides part of this one: NVIDIA GeForce RTX 4090 runs 8 of our 12 AI workloads while the other card runs 6, models that don't fit score zero.
| Benchmark | GeForce RTX 4080 Super | NVIDIA GeForce RTX 4090 | Difference |
|---|---|---|---|
| Qwen3 4B tok/s | 197.53 | 260.54 | -24% |
| Llama 3.1 8B tok/s | 126.29 | 171.29 | -26% |
| Qwen2.5-Coder 14B tok/s | 70.55 | 95.12 | -26% |
| Qwen3 32B tok/s | 0 | 44.28 | n/a |
| Llama 3.3 70B tok/s | 0 | 0 | n/a |
| Stable Diffusion XL images/min | 9.12 | 16.28 | -44% |
| Z-Image Turbo images/min | 0.75 | 7.425 | -90% |
| 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 | 1.42 | 4.7 | -70% |
| Wan 2.2 5B (720p) frames/s | 0 | 0.43 | n/a |
How long each card takes to finish a complete pipeline, not just one model. GeForce RTX 4080 Super is faster on 0 of 1; NVIDIA GeForce RTX 4090 on 1.
| Workflow | GeForce RTX 4080 Super | NVIDIA GeForce RTX 4090 | Difference | Cost per run |
|---|---|---|---|---|
| Full codebase review | 14.2 min | 10.5 min | NVIDIA GeForce RTX 4090 1.35x faster | $0.066 vs $0.027 |
Renting by the hour, NVIDIA GeForce RTX 4090 finishes 1 of 1 cheaper. The quicker card is not automatically the cheaper way to get the work done.
| Card | Per hour |
|---|---|
| GeForce RTX 4080 Super | $0.280 |
| NVIDIA GeForce RTX 4090 | $0.155 |
NVIDIA GeForce RTX 4090 is 1.81x cheaper per hour. Cheapest on-demand rate we see across RunPod and Vast.
| GeForce RTX 4080 Super | NVIDIA GeForce RTX 4090 | |
|---|---|---|
| VRAM | 16GB | 24GB |
| Transistors | 45,900M | 76,300M |
| Die size | 378.6 mm² | 608.5 mm² |
| Process node | 4 nm | 4 nm |
| Transistor density | 121.2 M/mm² | 125.4 M/mm² |
| Architecture | Ada Lovelace | Ada Lovelace |
| Memory bandwidth | 736 GB/s | 1008 GB/s |
| Boost clock | 2,505 MHz | 2,520 MHz |
| TDP | 320 W | 450 W |
| Launch MSRP | $999 | $1,599 |
| Release | 2024-01-31 | 2022-10-12 |
GeForce RTX 4080 Super full review · NVIDIA GeForce RTX 4090 full review · All AI & Machine Learning rankings