【WAN2.2】TXT to VIDEO — v2.1

【WAN2.2】TXT to VIDEO is a ComfyUI text-to-video workflow: enter a written prompt and load Wan Video 2.2 T2V-A14B to create video, with optional SeedVR2 upscaling, LoRA controls, and saved images containing generation parameters.

The workflow is described as a 12-node ComfyUI setup built with the UmeAiRT Toolkit.

The auto version includes automatic model downloads.

Use a written prompt, a loaded Wan model, and optional LoRA modules as inputs.

It creates video and saves images with generation parameters embedded.

Before loading it, enable "Use Nodes V2 (Vue)" in ComfyUI and refresh the page; display problems may occur without this setting.

Suggestion

Install ComfyUI-UmeAiRT-Toolkit through ComfyUI Manager by searching for "UmeAiRT," or use the UmeAiRT Auto-Installer.

Suggestion

The required custom node named by the workflow is ComfyUI-UmeAiRT-Toolkit.

The workflow names these model entries: WAN_2.2/T2V, 4x_NMKD-Siax_200k.pth, and seedvr2_ema_3b_fp8_e4m3fn.safetensors.

4x_NMKD-Siax_200k.pth is required by the workflow.

Turn SeedVR2 upscaling on or off.

Each of the three LoRA slots has its own enable toggle and strength control; additional LoRA modules can be connected.

No license is specified for this workflow.