Blog
Local AI animation: guides and real tests
Everything here was rendered and timed on a single consumer PC (RTX 5070 12GB, 64GB RAM).
GPU at 100% but Your AI Video Render Is Stuck? Check the Power Draw
Both renders showed 100% GPU. One drew 230 W and finished; the other drew 71 W and did 1 of 24 steps in 67 minutes. How to tell the difference on your own PC.
Wan 2.2 on an 8GB GPU? What Real Tests Show About VRAM and RAM
The GPU used only 8–9GB while the models need ~34GB, and a 16GB RAM limit didn't slow it down. Render times, memory use, and what you really need.
Turn Your Drawing into Animation with AI – Free and Local (Wan 2.2)
Keep the drawing style, restyle it, or turn it into something realistic. Step by step, with real results.
First Frame Last Frame AI (Free, Wan 2.2): 3 Runs with Unrelated Images
A sketch and a photo-real apple, same prompt, three runs, three different transitions.
Wan 2.2 on a 12GB GPU: Real Render Times and RAM Use (RTX 5070)
832×480 in 10.6 minutes, 1536×864 in 67 minutes, and why system RAM matters.
How to Animate a Drawing Locally: Loomance vs ComfyUI on WAN 2.2
Side-by-side test of a manual ComfyUI node graph and Loomance's preset workflows.