SDXL LoRA/LyCORIS works best on its original training model. You may need to reduce weight when using it with other models or additional LoRAs.
LyCORIS Combination Recommendation: Try using four of my LyCORIS models with the following weights: 0.4, 0.3, 0.2, 0.2 (or any combination with a total weight around 1.0).
I will release multiple versions for a few popular models. Feel free to request for artist or model.
Version naming convention: MajorVersion.MinorVersion [BaseModel]
MajorVesion: For big updates that can apply to any style LoRA. For example, training parameter update.
MinorVersion: For small updates that only apply to this LoRA. For example, epoch selection, removing a certain image from dataset.
BaseModel: SDXL LoRA/LyCORIS works best on the model that it was trained on.
Description
FAQ
Comments (2)
I was initially getting "nonetype not iterable" errors in reForge when trying to use this model because it's tagged as a LyCORIS, so CivitAI Helper put it in the LyCORIS folder and webui was auto-populating "<lyco:...>" instead of "<lora:...>".
The fix was to move it to the Lora folder and use "<lora:...> instead of "lyco:...>".
Not sure if this was just an issue on my end (I have the sd-webui-lycoris extension installed), but if anyone else runs into this, put it in the Lora folder and call it out as a Lora.
If you normally put all Lora and LyCORIS models in a single Lora folder then you probably won't ever run into this type of issue. Seems like a problem with how reForge interacts with LoKr files like this one.
Thanks for the awesome model (and the huge gallery of examples to go with it)!
IIRC, the "<lyco:...>" method has been obsolete for while now, and it is now common practice to put all lora-types in a single folder, and use the "<lora:...>" method, without needing any extensions.
EDIT: Yes, just checked in A1111, it has been supported since v1.5.0 released in Jul 2023.
I don't know about reForge, but if it's a fork of A1111, then I'm guessing it should work like that too.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.