main
feature: Add DiT model support with toroidal attention, latent wrapping, and Raylight integration
ComfyUI Advanced Tiling
ComfyUI tiling nodes inspired by spinagon/ComfyUI-seamless-tiling. This project enables the creation of tileable images in various shapes with customizable rotations. Example workflows are located in the workflows directory.
⚠️ There are some artifacts present. I'm not sure if this is because of a flawed implementation or simply because Stable Diffusion isn't intended for this purpose.
Implemented tiling modes
- Hexagon
- Rectangular (toroidal — right edge wraps to left, bottom wraps to top)
- None (normal generation)
Supported models
UNet-based
- Stable Diffusion 1.5 (also 1.4)
- Stable Diffusion 2.1 (also 2.0)
- Stable Diffusion XL (SDXL)
DiT-based (via toroidal attention patching)
- FLUX.2 — tested on flux.2-klein-4b
- Qwen Image — tested on qwen-image-2512 (Q6 GGUF and fp16)
DiT models are automatically detected and use toroidal attention instead of latent padding for seamless tiling.
TODO
- Optimize VAE decode (first pass is very slow)
Credits
spinagon/ComfyUI-seamless-tiling [GPL-3.0] - Used as a base for this project
Red Blob Games - Hexagonal Grids - Hexagonal grid math
Languages
Python
100%
