These are estimates. Real results vary ±10–20% based on your CPU, background processes, and optimization flags. Using --use-sage-attention in ComfyUI adds ~15–20% speed on RTX 30–50 series. CPU speed and system RAM have minimal effect (<10%) on GPU-bound SDXL generation — your GPU is almost the entire story. Benchmarks sourced from Chimolog (WAI-NSFW v14, 832×1216, 28 steps, ComfyUI) and community testing.

Wan 2.2 key requirements: Minimum 64GB system RAM recommended (model offloads to RAM during generation — 32GB causes SSD spillover and dramatic slowdowns). 720p requires VRAM 16GB+. Only GeForce GPUs run 720p reliably — AMD and Intel Arc fail with OOM errors at that resolution. Using the Kijai/EasyWan22 workflow with Lightx2v LoRA (8 steps). Benchmarks from Chimolog Wan 2.2 benchmark · 81 frames · ComfyUI.