WanGP v18.21 "No GPU Left Behind": auto memory mode, faster RTX 20xx, FastH3 Omni and Fibo 2.0
WanGP, the low-VRAM app for running open video and image models locally, added an automatic memory profile and big speedups for older RTX 20xx cards. The same update adds FastH3 Omni, Fibo 2.0 with Edit Turbo and the FlashVSR+ Turbo upscaler.
What it does
Profile 0 picks RAM and VRAM settings for your machine automatically. With WGPKernels 1.3.0, LTX-2.5 VAE encode, decode and upscale run 4.8 to 6.8 times faster on RTX 20xx, and most models run up to 10% faster. FastH3 Omni needs an RTX 30xx or newer and has an NVFP4 variant; H3 two-pass runs get a separate Phase 2 Steps setting.
Worth knowing
FlashVSR+ Turbo upscales 1 to 4 times, and a new --offline flag stops it from reaching the internet. The built-in Deepy Prime assistant with Qwen 3.8 Bonsai runs at 60 tokens a second on a 2080 Ti. MiniMax H3: Users in the EU, UK, US and South Korea need to request a license from MiniMax via the H3 License Request Form (platform.minimax.io/h3-license) before use.