scraed/LanPaint

High quality training free inpaint for every stable diffusion model. Supports ComfyUI

52
/ 100
Established

Implements an asymptotically exact conditional sampling algorithm that enables iterative "thinking" refinement steps during the diffusion process, allowing models to reconsider inpainted regions before final denoising. Supports a wide range of architectures including Flux, SDXL, SD 3.5, Hunyuan, Wan 2.2, Z-image, and Qwen models, with native ComfyUI integration as a drop-in KSampler replacement and parallel Diffusers library support for broader framework compatibility.

1,050 stars. Actively maintained with 5 commits in the last 30 days.

No Package No Dependents
Maintenance 13 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 13 / 25

How are scores calculated?

Stars

1,050

Forks

34

Language

Python

License

GPL-3.0

Last pushed

Mar 03, 2026

Commits (30d)

5

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/scraed/LanPaint"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.