Sign In

Image Cleanup Studio — Free Lite Text-to-Cutout (WeVisOne-VL) — ComfyUI

Download

1 variant available

Archive Other

Image_Cleanup_Studio_Pack_Civitai_Free.zip

2.15 KB

Verified:

Type
Workflows
Stats

17

Reviews

No reviews yet

Published

Nov 17, 2026

Base Model

Other

Hash
AutoV217E6477C21
cover.jpg

FREE starter on Whop (lite workflow + exclusive cheat sheet): https://whop.com/imagelab/free-starter-image-cleanup-studio

FULL pack ($39): https://whop.com/imagelab/image-cleanup-studio-cutouts-object-removal-photo-deblur-comfyui — use code CLEAN30 for 30% off ($39 → $27.30)

## FREE vs FULL

| | FREE (this download) | FULL ($39) |

|---|---|---|

| Workflows | 1 — WeVisOne-VL cutout | 4 — cutout + remove-object + deblur compare + restore chain |

| Remove-Object 6-step | — | Yes |

| PhotoDeblur (HI-Diff + MISCFilter) | — | Yes |

| Restoration chain + 2K upscale | — | Yes |

| Prompt library | — | Cutout queries + removal instructions + deblur recipes |

| Guides | This README | Full EN quickstart + Arabic guide |

## What this is

WeVisOne-VL (Tencent, Apache 2.0) turns a text phrase into a pixel-perfect mask — cut anything out of a photo with one sentence, with per-instance alpha mattes and a transparent PNG out. This free lite workflow is the core cutout graph, running on ComfyUI's native core node (0.45.0+).

## Files you need (not included)

From Comfy-Org/WeVisOne-VL: wevisone_vl_4b_int8_convrot.safetensors (5.3 GB) → text_encoders. Load via CLIPLoader, type stable_diffusion. ~16 GB VRAM.

## License

WeVisOne-VL (Apache 2.0) and PhotoDeblur (MIT) are commercial OK. Remove-Object outputs in the full pack are NON-commercial (Qwen-Image 2.1 Research License).