As the model is currently unavailable in the CivitAI generator, you can use it on sea.art: HERE
This model is a merge of PonyDiffusionV6XL oriented more towards anime illustrations and the Mistoon aesthetic. To get the most out of it, be sure to include the following string in the positive prompt:
score_9, score_8_up, score_7_up,score_6_up
There are two different version of this model right now:
1.0: This is the standard one, use this if you are a beginner
1.0 Fast: This is merged with SDXL Lightning to improve generation speeds. If you are going to use this version I suggest using Euler A SGMUniform with at least 8 steps and a CFG of 4.
2.0 Fast: A new version of the model oriented towards Anime illustrations. This use the same settings of 1.0 Fast but with a CFG of 2.5
1.0 Hyper: This is the fastest version (but with the lowest quality). It is merged with the Hyper LoRA to achieve improved generation speed. When using this be sure to use DPM++ SDE Karras with 4-6 steps and 2/2.5 CFG Scale.
As this is a merge of that awesome model the same license applies:
License
This model is licensed under a modified Fair AI Public License 1.0-SD (https://freedevproject.org/faipl-1.0-sd/) license.
The following modifications have been added to Fair AI Public License:
You are not permitted to run inference of this model on websites or applications allowing any form of monetization (paid inference, faster tiers, etc.). This applies to any derivative models or model merges.
If you want to use this model commercially, please reach us at [email protected].
Explicit permission for commercial inference has been granted to CivitAi and Hugging Face.
Learn more about how I create the examples here: From Noise to Illustrations: How I Generate AI Pictures with Stable Diffusion | Civitai
In the past few months I've had a lot of people asking me why their pictures are not as detailed or clean as mine. I've written a few articles about SD, but if you want to learn my exact workflow I highly recommend you to check out my latest one: My Workflow
If you want to test this (or other models), I have made an extension to help me create pictures using a variety of different tags. The previews images have been generated using my custom extension Ranbooru which scrapes tags from random images on gelbooru, safebooru and rule34.
If you want to learn more about my workflow, check out the guides below.
Learn how to create pictures like mine with my step-by-step tutorials:
Other stuff here on CivitAI:
I've got a lot of models here on CivitAI (and even more on Patreon), here's some of them:
Initially I used to release every LoRA here on CivitAI, but after a while it became very time-consuming, so I started releasing only my favourite new LoRAs here on CivitAI while still releasing all the models for free on Patreon.
So, if you don't want to miss a release be sure to follow me on Patreon (for free you don't need to sub if you don't want to) here:
https://www.patreon.com/Inzaniak
If you want to check out my other models you'll find them here:
Civitai: Civitai | Share your models
Patreon (Free): Inzaniak | Generative Art and Tech Blogging | Patreon
Patreon (All): Inzaniak | Generative Art and Tech Blogging | Patreon
I've started developing custom models for myself a few months ago just to check out how SD worked, but in the last few months it has become a new hobby I like to practice in my free time. All my checkpoints and LoRAs will always be released for free on Patreon or CivitAI, but if you want to support my work and get early access to all my models feel free to check out my Patreon:
https://www.patreon.com/Inzaniak
If you want to support my work for free, you can also check out my music/art here:
Description
This is a merge of the SDXL-Lightning LoRA with my Mistoon_Copper_XL model
FAQ
Comments (12)
I can't seem to find your recommended Euler A SGMUniform anywhere. If you could point me in the right direction it would be much appreciated.
It is usually in ComfyUI. KSamplers have sampler name and scheduler as inputs. Euler A is a sampler name, SGMUniform is a scheduler.
Hey! I'm using the SD UI Forge version which comes with a few added schedulers (like the Euler A SGMUniform). I highly recommend to check it out as it boosts the generation speed and improves memory efficiency (lllyasviel/stable-diffusion-webui-forge (github.com)). You should also be able to use the Euler A Turbo scheduler with comparable results.
Hmmmm, model is cool and all but faces don't really look like mistoon model at all. (which is sad cuz i love mistoon for it's faces actually)
I noticed I wasn't able to find this checkpoint when looking for SDXL Lightning, and see you have it listed under Turbo instead. Would this not quality for the lightning category for checkpoints?
Yep, my bad. I didn't notice that there is a SDXL Lighting tag. Now it should be correct.
Why do all your examples for Lighning-derived models have such enormous number of steps?
They should be tested at 2-8 steps, not 40!
Hey! I'm using the Forge version of the Automatic1111 webui. These pictures have been created starting with 8 steps at 680x1024 then img2img to 1024x1536 and finally img2img to 1280x1920. The reason you are seeing 40 steps is actually due to the fact that the UI is configured to run with 40 steps in img2img, but it's not actually running that number of steps. Forge automatically reduces the amount of steps needed based on the denoising strength in the img2img page, so it's actually doing 4 steps while showing 40 in the metadata.
What does score_6_up bring?
It's a tag used to improve output quality in Pony Diffusion. You can learn more about it in the original model page: Pony Diffusion V6 XL - V6 (start with this one) | Stable Diffusion Checkpoint | Civitai
Ok. I asked because most articles suggest putting score_6 in negative.
So I assume that those scores are equivalent to the following quality tags:
score_9 = masterpiece,
score_8 = best quality
score_6 = highly detailed
@MangousteAI For XL it is recommended to use the full string in the positive prompt as described in the checkpoint post:
score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up
It says that it's due to an issue that occurred during the training process.
Details
Files
Available On (3 platforms)
Same model published on other platforms. May have additional downloads or version variants.