multidiffusion-upscaler-for-automatic1111 VS loopback_scaler

Use tiled VAE, it will save VRAM: pkuliyi2015/multidiffusion-upscaler-for-automatic1111: Tiled Diffusion and VAE optimize, licensed under CC BY-NC-SA 4.0 (github.com)

My favorite solution is to not use ultimate Sd upscale and instead use multidiffusion-upscaler.

sure, i personally use 'tiled diffusion' https://github.com/pkuliyi2015/multidiffusion-upscaler-for-automatic1111, works like charm also use adetailer for faces if its needed.

What worked for my 1080 was using TiledVAE and turning down the quality of my previews - I don't pay much attention to it/s but it's definitely faster than using --medvram, and now I can handle batches and large resolutions without things exploding on me.

I would highly recommend adding https://github.com/pkuliyi2015/multidiffusion-upscaler-for-automatic1111 to your A1111 installation, TiledVAE is enabled automatically under the hood in A8R8; this will allow you to get even larger generations before getting an out of memory error. You'll get a Tiled Diffusion checkbox with some reasonable hardcoded defaults as well.

Use Tiled VAE https://github.com/pkuliyi2015/multidiffusion-upscaler-for-automatic1111

That's weird. 10GB should allow you to hires to 2048x2048 at least. Use Tiled VAE extension https://github.com/pkuliyi2015/multidiffusion-upscaler-for-automatic1111 that will allow you to go even beyond that.

There are multiple ways if you are using SD Web UI. One is to use loopback scaler as I did: https://github.com/Elldreth/loopback_scaler

https://github.com/Elldreth/loopback_scaler this really makes images cool

This upscaler works really well https://github.com/Elldreth/loopback_scaler it adds a lot of details

Extensions: https://github.com/Mikubill/sd-webui-controlnet https://github.com/Elldreth/loopback_scaler https://github.com/hako-mikan/sd-webui-regional-prompter (used only on one image)

Loopback Scaler: https://github.com/Elldreth/loopback_scaler

Loopback scaler is really good at adding detail to the image. Essentially it does img2img (and some other tweaks) over and over in a loop. It gradually makes the image larger between the loops, but there's a limit (limited by VRAM) of how big of an image it can generate. On the cons side it can add unwanted artifacts, just like when you generate images much larger than 512x512 in txt2img/img2img.

Is this the Loopback Scaler you were talking about? I gotta give that a try.