Sable Diffusion版本v1.0 (ID: 453793) 综合资源合集综合资源合集

I am admittedly kind of a noob at this, but if you enjoy my humble work, any donations are greatly appreciated: https://ko-fi.com/insult_to_ninjary

Your generosity helps me stay motivated to keep improving and to keep trying new things!

The name kind of writes itself for this one, doesn't it? lol

As you may have guessed, this LoRA will give your art an aesthetic similar to the video game, Sable. It didn't come out as crisp as I had hoped, but I still think it does a decently good job at capturing the vibe of the game. It's also possible I just haven't discovered the best base model+prompt combo yet for this LoRA. If you find any setup that works well, let me know about it!

Here are some tips for using the LoRA in the meantime:

1. Illustrated base models will get you closest to the game's art style, but they seem to struggle at reproducing the fine lines and details that the game displays. 2.5D models do a bit better at detailing, but lose accuracy to the game's art style (pics still look nice tho). Realistic models look great, but only have a loose resemblance to the game. I recommend experimenting a lot, and let me know if you find any parameters that work well!

2. If you're having problems with anatomy or with getting your desired image composition, you can try gradually lowering the weight of the LoRA. I always recommend starting at a weight of 1 and then gradually adjusting up or down as necessary (probably down in this case).

3. Use a VAE, especially if your chosen base model doesn't have one baked in! It really seems to help with getting better colors and anatomy.

4. Upscale or use Hires. fix with a quality upscaler. If Hires. fix seems to be destroying your images, try lowering the denoising strength.

If you make anything awesome, please consider posting them to the gallery! That goes for all of my LoRAs! I really like seeing what kind of stuff you all are able to make!

If you have any questions, feel free to ask! I will try my best to help! Also, if you have any ideas for future LoRAs, please don't hesitate to share them with me!