Trained on Illustrious-XL.
gideon:
https://twitter.com/gideonscorral
http://www.furaffinity.net/user/gideon
https://www.patreon.com/gideonscorral
Prompt :
masterpiece, best quality, amazing quality, 91d30n, optional: 2018, 2019
Negative prompt :
monochrome, greyscale, signature, url, artist name, username, text, watermark, English text, bad quality, worst quality, worst detail, sketch, censor, Description
Trained on Illustrious-XL(v1.1). { "engine": "kohya", "unetLR": 1, "clipSkip": 2, "loraType": "lora", "keepTokens": 1, "networkDim": 16, "numRepeats": 2, "resolution": 1024, "lrScheduler": "cosine", "minSnrGamma": 5, "noiseOffset": 0.1, "targetSteps": 4225, "enableBucket": true, "networkAlpha": 8, "optimizerType": "Prodigy", "textEncoderLR": 1, "maxTrainEpochs": 19, "shuffleCaption": true, "trainBatchSize": 6, "flipAugmentation": false, "lrSchedulerNumCycles": 3 }
FAQ
Comments (3)
Whats difference between this model and previous one? Its looking like previous version is better then new one
The main difference is in the base models used during training - this one uses Illustrious 1.1 while the previous version was based on 0.1. Some users might find the 0.1 version performs better due to broader compatibility, though personally I think the difference is minimal. I released this update primarily because many people use this model, and wanted to provide more options for the community.
This lora works amazing for hyper proportions even at low strength. Really great, thanks!