Simple WanVideo ComfyUI workflows (tnx kijai) — Image2video Wan2.1 v1.0

Use this ComfyUI workflow to turn text into images with simple upscaling, or images into videos with prompt enhancement.

For text-to-image, provide text; the workflow describes an image output with simple upscaling.

For image-to-video, provide an image; the workflow describes a video output with prompt enhancement.

Find the workflow on Civitai: https://civitai.com/models/1299250.

Required components listed for the workflow include ComfyUI-WanVideoWrapper, ComfyUI-Easy-Use, ComfyUI_LayerStyle, ComfyUI-utils-nodes, and ComfyUI-Custom-Scripts.

Named model files include MiaoshouAI/Florence-2-base-PromptGen-v2.0, open-clip-xlm-roberta-large-vit-huge-14_fp16.safetensors, wan_2.1_vae.safetensors, and Wan2_1-I2V-14B-480P_fp8_e4m3fn.safetensors.

Named workflow nodes include WanVideoSampler, VHS_VideoCombine, DownloadAndLoadFlorence2Model, LoadWanVideoT5TextEncoder, TextPreview, WanVideoImageClipEncode, WanVideoBlockSwap, WanVideoModelLoader, and WanVideoDecode. Other named nodes are LoadWanVideoClipTextEncoder, easy prom

Choose the text-to-image workflow for text input, or the image-to-video workflow for image input.

Vorschlag · nicht geprüft

The Civitai listing does not state a license; check its terms before reusing the workflow.

Vorschlag · nicht geprüft