WebUI runs LTX2.3, Wan2.2, Flux.2 on 6G/8G VRAM

LiteUI-Studio uses a ComfyUI backend to run quantized GGUF models (LTX2.3, Wan2.2-A14B, Flux.2-Klein-9B) on 6GB/8GB VRAM. Supports loading finetuned models and LoRAs, with no node editing required.