Confused about the different model versions? Check the 'About this version' on the right sidebar for more information.
This is just a quick release that we did since Flux + SD3.5 is taking such a long time (but we are working on it!)
Hopefully this satisfies those furry desires.
Take care!
Model Info
Chroma XL is an experimental model that is a partial merge + train of the brilliant NoobAI-XL (NAI-XL) model by L_A_X and their team, NoobAI is based off of Illustrious. All credit to the model goes to him.
The model only needs low CFG and low step count. Making it highly versatile over a wide range of consumer hardware.
We thank L_A_X and their team for his fantastic efforts in creating this model!
Recommended Settings
We recommend using the following settings with the model for best results.
Resolution: 1024x1024
Steps: 8-12
CFG Guidance: 1.7-3.0
Sampler: Euler A
Scheduler: beta
The model utilises the low guidance and steps. It's advised to keep your step count as well as your guidance low, whilst you may increase the step count if you choose to do so, there is a clear point of diminishing returns, therefore we recommend you to follow our recommended values here.
It is strongly recommended to use an ancestral sampler (e.g. Euler a) with beta schedule, the noise injection in ancestral sampler is absolutely needed to prevent over saturation.
Usage Instructions
All the sample images have their workflow included within them. It’s recommended to download those images and use that workflow in ComfyUI.
For all other clients, just run it as you would with any other SDXL model.
Strengths & Weaknesses
During our evaluations of the model, we find the following strengths and weaknesses in the model.
Strengths
High-level understanding of danbooru and e6 tags.
Relatively good anatomy compared to previous models.
Good understanding of many booru concepts, especially ones related to anime and furry.
Ability to do duo quite coherently, especially compared to prior models.
Ability to do high quality backgrounds.
Weaknesses
Non-existent natural language capabilities (don’t even try to use it).
Slightly incompatibility with SDXL model extensions, such as ControlNet but give it a go.
Characters that are in the distance tend to become extremely blurry.
The whole model has sort of a burned in ‘haze’ which can cause upscaling to look extremely pixelated.
Can look overcooked sometimes.
Prompting & Tips
The model has next to no knowledge of natural language so you should exclusively use tags with this one.
It’s recommended to go over the guidelines provided within the base models CivitAI page here.
https://civarchive.com/models/833294?modelVersionId=998979
The model was trained with artist tags.
It’s recommended to always include solo, duo, trio, etc. Since not adding any of those tags results in a very unpleasant amalgamation of them.
Credits
This model was merged + trained by the Horizon Team.
All credit for the base model goes to L_A_X and their team.
https://civarchive.com/models/833294?modelVersionId=998979
Contact Us
You can reach us at our Discord server!
Description
FAQ
Comments (15)
Since this is based on NoobAI-XL, which in turn is based in Illustrious, please change base model type from "SDXL 1.0" to "Illustrious".
Not sure what you mean by this? Illustrious is a finetune of SDXL and NoobAI-XL is a finetune on top of that. They shouldn't really be considered base models. We just call them that since it's easier to understand. But from a technical perspective, they are all based on SDXL architecturally.
@HorizonTeam on civitai they added illustrious as a base model type so that people can figure out stuff like what loras are more likely to be compatible with each other
@HorizonTeam By correctly attributing your model, you'll make it much easier to find for people who filter by tags. Same thing with Pony, which is technically also an SDXL model, but has its own base model type on this site for convenience.
Model was updated to have Illustrious as the base model after tests of some LoRA's proved decent compatibility.
It's fine to keep as Illustrious, as it is more accurate but as a slight counter argument: hardly anybody knows what Illustrious is and now people looking for XL compatible models that apply a filter on XL will no longer find this one. Feels like this will hinder discovery.
@renfald Yeah, I didn't even know what Illustrious was until I saw NoobAI. Seems to be popular in anime circles but I don't really bother with those. I suppose an alternative could be to upload the same model as a different version and set the base model to SDXL, which would be able to be discovered by both. Not sure if I want to do that though. CivitAI is just bad in general at discovery. I don't know how well SDXL LoRA's will work though.
I think there is a greater number of people searching for SDXL models compared to Illustrious models. So, I think for discovery, I might change it back to SDXL. It's a flaw of CivitAI that we cannot have both I'm afraid.
EDIT: Model was changed back to SDXL 1.0 as it seems more likely that people will be searching for an SDXL model rather than an Illustrious model. Illustrious LoRA's can be searched regardless.
@HorizonTeam Perhaps mention Illustrious prominently near the top of the model description as I kind of middleground?
Where is Scheduler: beta?
It should be there by default. I can see it here on Reforge.
make sure your UI is up to date
Update your UI, its somewhat new.
Does the model have some "accelerator" lora merged in?
Not sure on the specifics but I believe this model is similar to CompassMix https://civitai.com/models/498370/compassmix-xl-lightning which uses Lightning in addition to Lodestone's "magic sauce" which I think is something with GANs
@CosmicElement its a good merge but the last 3 weaknesses is probably caused because of the lightning part of the merge.