Spaces:

---

ltx-community

/

ltx-2.3-ingredients-distilled

like 32

Running on Zero

Fuses the LoRA + loads the precompiled AOTI transformer blocks at the root module level (per ZeroGPU docs). LoRA is fused & unloaded (weight names match the compiled constants); the runtime scale slider is preserved by re-fusing the delta + re-snapshot only on change (validated on deblur). ~1.3x faster.