A highly efficient, secure, and stable application development platform with excellent performance, easy scalability, and deep integration of AI capabilities such as LLM, AI Chat, RAG, and Agents.
🪝 Claude Code hooks + an installable plugin marketplace: safety, cost, observability, productivity.
关键参数与规格
关键参数与规格
参数
值
语言
Python
类别
AI Engineering
子类别
Build agent workflow (visual)
关键词
Diffusion, Image production
星标数
544
分叉数
66
贡献者
2
创建日期
2024-01-07
更新日期
2026-09-19
1 天星标数
0
7 天星标数
0
排名
17,349
上次排名
17,135
排名变化
-214
7 天星标排名
16,059
上次 7 天星标排名
16,066
7 天星标排名变化
7
总星标排名
18,141
上次总星标排名
18,117
总星标排名变化
-24
README 摘录
Tiled Diffusion & VAE for ComfyUI Check out the SD-WebUI extension for more information. This extension enables **large image drawing & upscaling with limited VRAM** via the following techniques: Reproduced SOTA Tiled Diffusion methods pkuliyi2015 & Kahsolt's Tiled VAE algorithm ~~pkuliyi2015 & Kahsolt's TIled Noise Inversion method~~ [! NOTE] Sizes/dimensions are in pixels and then converted to latent-space sizes. Features [x] Supported models [x] ControlNet support [ ] ~~StableSR support~~ [ ] ~~Tiled Noise Inversion~~ [x] Tiled VAE [ ] Regional Prompt Control [x] Img2img upscale [x] Ultra
-Large image generation Tiled Diffusion [! TIP] * Set to 0 and to 1 to see the tile seams and then adjust the options to your needs. * Increase to increase speed (if your machine can handle it). * Use the colorfix node if your colors look off. | | | Tile's width | | | Tile's height | | | Tile's overlap | | | The number of tiles to process in a batch | How can I specify the tiles' arrangement? If you have the Math Expression node (or something similar), you can use that
to pass in the latent that's passed in your KSampler and divide the / by the number of rows/columns you want. = number of columns you want = number of rows you want = = SpotDiffusion Paper A tiling algorithm that attempts to eliminate seams by randomly shifting the denoise window per timestep. It is mainly used for fast inferences by setting to 0; otherwise, it's better to stick with the other tiling strategies as they produce better outputs. This additional feature is experimental
, in testing, and subject to change. Tiled VAE The recommended tile sizes are given upon the creation of the node based on the available VRAM. NOTE] Enabling for the decoder may produce images with slightly higher contrast and brightness. | | | When Fast Mode is disabled: The original VAE forward is decomposed into a task queue and a task worker, which starts to process each tile. When GroupNorm is needed, it suspends, stores current GroupNorm mean and var, send everything to RAM
, and turns to the next tile. After all GroupNorm means and vars are summarized, it applies group norm to tiles and continues. A zigzag execution order is used to reduce unnecessary data transfer. When Fast Mode is enabled: The original input is downsampled and passed to a separate task queue. Its group norm parameters are recorded and used by all tiles' task queues. Each tile is separately processed without any RAM-VRAM data transfer. After all tiles are processed, tiles
are written to a result buffer and returned. | | | Only estimate GroupNorm before downsampling, i. Can restore colors if tiles are too small. | Workflows The following images can be loaded in ComfyUI. License Great thanks to all the contributors! 🎉🎉🎉 The implementation of MultiDiffusion, Mixture of Diffusers, and Tiled VAE code is currently under Creative Commons Attribution-NonCommercial-ShareAlike 4. 0 International License since it was borrowed from the wonderful SD-WebUI extension.
If you have the Math Expression node (or something similar), you can use that to pass in the latent that's passed in your KSampler and divide the tile_height/tile_width by the number of rows/columns you want.
C = number of columns you want R = number of rows you want
pixel width of input image or latent // C = tile_width pixel height of input image or latent // R = tile_height
A tiling algorithm that attempts to eliminate seams by randomly shifting the denoise window per timestep. It is mainly used for fast inferences by setting tile_overlap to 0; otherwise, it's better to stick with the other tiling strategies as they produce better outputs.
This additional feature is experimental, in testing, and subject to change.
Tiled VAE
The recommended tile sizes are given upon the creation of the node based on the available VRAM.
[!NOTE]
Enabling fast for the decoder may produce images with slightly higher contrast and brightness.
Options
Name
Description
tile_size
The image is split into tiles, which are then padded with 11/32 pixels' in the decoder/encoder.
fast
When Fast Mode is disabled:
The original VAE forward is decomposed into a task queue and a task worker, which starts to process each tile. When GroupNorm is needed, it suspends, stores current GroupNorm mean and var, send everything to RAM, and turns to the next tile. After all GroupNorm means and vars are summarized, it applies group norm to tiles and continues. A zigzag execution order is used to reduce unnecessary data transfer.
When Fast Mode is enabled:
The original input is downsampled and passed to a separate task queue. Its group norm parameters are recorded and used by all tiles' task queues. Each tile is separately processed without any RAM-VRAM data transfer.
After all tiles are processed, tiles are written to a result buffer and returned.
color_fix
Only estimate GroupNorm before downsampling, i.e., run in a semi-fast mode.
Only for the encoder. Can restore colors if tiles are too small.
@article{jimenez2023mixtureofdiffusers,
title={Mixture of Diffusers for scene composition and high resolution image generation},
author={Álvaro Barbero Jiménez},
journal={arXiv preprint arXiv:2302.02412},
year={2023}
}
@article{bar2023multidiffusion,
title={MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generation},
author={Bar-Tal, Omer and Yariv, Lior and Lipman, Yaron and Dekel, Tali},
journal={arXiv preprint arXiv:2302.08113},
year={2023}
}