【Anima】INPAINT — v1.0

This ANIMA Inpaint workflow from UmeAiRT is for inpainting and text-to-image generation in ComfyUI; it takes an ANIMA model and text prompt, then produces saved images with optional SeedVR2 upscaling and generation parameters embedded in the image metadata.

The described eight-node flow loads the model, accepts the prompt, generates and saves images, and includes optional upscaling and metadata embedding.

Saved images include all generation parameters in their metadata for online publishing and remixing.

Manual setup identifies an ANIMA model, qwen_3_06b_base.safetensors text encoder, and qwen_image_vae.safetensors VAE as files used by the workflow.

Turn on ComfyUI's "Use Nodes V2 (Vue)" setting, refresh the interface, and then load the workflow.

Suggestion · non vérifié

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

Suggestion · non vérifié

Before use, verify that ANIMA/Base, seedvr2_ema_7b_sharp_fp8_e4m3fn_mixed_block35_fp16.safetensors, and ComfyUI-UmeAiRT-Toolkit are available.

Suggestion · non vérifié

Manual setup lists these files: anima-base-v1.0.safetensors as the base model, qwen_3_06b_base.safetensors as the text encoder, and qwen_image_vae.safetensors as the VAE.

For the GGUF version, the listed model options are Q8, Q6, Q5, Q4, and Q3 in ComfyUI/models/unet.

Three LoRA slots have individual on/off toggles and strength controls, and additional LoRA modules can be connected.

SeedVR2 upscaling can be toggled on or off; the description characterizes it as slower than a classic upscaler but with significantly better quality.

The auto version is described as downloading models automatically.

Without enabling ComfyUI's Nodes 2.0 (Vue) interface, the workflow may have display problems.

No license is listed for this workflow.