

NVIDIA GeForce RTX 4090 wins 8 of 12 benchmarks, averaging 168% 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 350% (7.425 vs 1.65 images/min); the closest fight is Qwen3 4B (93% 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 | NVIDIA GeForce RTX 4090 | GeForce RTX 5060 Ti | Difference |
|---|---|---|---|
| Qwen3 4B tok/s | 260.54 | 134.96 | +93% |
| Llama 3.1 8B tok/s | 171.29 | 84.2 | +103% |
| Qwen2.5-Coder 14B tok/s | 95.12 | 45.06 | +111% |
| Qwen3 32B tok/s | 44.28 | 0 | n/a |
| Llama 3.3 70B tok/s | 0 | 0 | n/a |
| Stable Diffusion XL images/min | 16.28 | 5.22 | +212% |
| Z-Image Turbo images/min | 7.425 | 1.65 | +350% |
| 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 | 4.7 | 1.97 | +139% |
| Wan 2.2 5B (720p) frames/s | 0.43 | 0 | n/a |
How long each card takes to finish a complete pipeline, not just one model. NVIDIA GeForce RTX 4090 is faster on 1 of 1; GeForce RTX 5060 Ti on 0.
| Workflow | NVIDIA GeForce RTX 4090 | GeForce RTX 5060 Ti | Difference |
|---|---|---|---|
| Full codebase review | 10.5 min | 22.2 min | NVIDIA GeForce RTX 4090 2.11x faster |
| NVIDIA GeForce RTX 4090 | GeForce RTX 5060 Ti | |
|---|---|---|
| VRAM | 24GB | 16GB |
| Transistors | 76,300M | 21,900M |
| Die size | 608.5 mm² | 181 mm² |
| Process node | 4 nm | 4 nm |
| Transistor density | 125.4 M/mm² | 121 M/mm² |
| Architecture | Ada Lovelace | Blackwell |
| Memory bandwidth | 1008 GB/s | 448 GB/s |
| Boost clock | 2,520 MHz | 2,572 MHz |
| TDP | 450 W | 180 W |
| Launch MSRP | $1,599 | $329 |
| Release | 2022-10-12 | 2025-04-16 |
NVIDIA GeForce RTX 4090 full review · GeForce RTX 5060 Ti full review · All AI & Machine Learning rankings