Wan2.1 Vace T2I_Inpainting&out_gguf_8g_LowVram_feet&hand_fix_easy — V1

Use Wan2.1 Vace T2I_Inpainting&out_gguf_8g_LowVram_feet&hand_fix_easy for single-image inpainting and outpainting: provide one picture and a mask, then save the edited image.

It includes two workflow files: Wan2.1_Vace_T2I_inpainting_outpainting_gguf_LowVram_easy_V1.json and Wan2.1_Vace_T2I_inpainting_outpainting_gguf_LowVram_easy_V1 (add_Nag).json.

Named input components include LoadImage and MaskToImage.

Named image stages include MaskPreview+, Image Comparer (rgthree), and SaveImage.

The listing identifies this as a Workflows item from SamLiu and names Wan Video 14B i2v 480p as its base model.

The named model files are umt5_xxl_fp8_e4m3fn_scaled.safetensors, wan_2.1_vae.safetensors, and Wan2.1_14B_VACE-Q4_0.gguf.

Image and mask handling names include LoadImage, ImagePadForOutpaint, ImageCompositeMasked, MaskToImage, ImageResizeKJv2, ImageGenResolutionFromImage, MaskPreview+, ResizeMask, INPAINT_MaskedFill, VAELoader, VAEDecode, and SaveImage.

Other named components include Power Lora Loader (rgthree), CLIPLoader, CFGZeroStar, CLIPTextEncode, SkipLayerGuidanceDiT, Image Comparer (rgthree), ModelSamplingSD3, KSamplerWithNAG, MarkdownNote, UnetLoaderGGUF, UNetTemporalAttentionMultiply, and WanVaceToVideo.

Both width and height need to be divisible by 32; other values are described as causing an error.

For generation steps, try 4 for a preview, 8 or 12 for “Barely,” and at least 16 for “Basically the best it can do.”

Suggestion · non vérifié

Before use, verify that your setup includes the named model files and components.

Suggestion · non vérifié

The listing does not specify a license.