<h1 id="the-style-of-hairu.">The style of Hairu.</h1>I made this lora to try and reproduce the style of Hairu.ILLUSTRIOUS 2+I made a newer version of this lora, because I found the version before this one lacking.The new version has a better understanding of the style and also a bigger dataset for better range.I changed the trigger to just <code>hairu,</code> for better results.In the demo images it says the lora is version 2, but I later changed that during uploading to version 3 in the name. Please make sure to take this change in mind.Ok have fun with this new improved version.Trigger: <code>hairu,</code>Helpful tags: <code>masterpiece, best quality, traditional media, jaggy lines,</code>PONY / ILLUSTRIOUS 0.1The above showcase images were made with Addetailer extension enabled and using a face, eyes and hand model.This LoRa is trained on Pony V6 and Illustrious 0.1, but I made the above showcase images in different checkpoints, because their style was more to my liking.The Trigger is: <code>ha1ru,</code>Advised strength: 0.6-0.9Prompting IllustriousFor all versions I recommend you use: <code>masterpiece, best quality, amazing quality, very aesthetic, high resolution,</code> in your prompt and I myself am using the following negative prompt: <code>text, abstract, glitch, deformed, mutated, ugly, disfigured, long body, lowres, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, lowres, (worst quality, bad quality:1.2), bad anatomy, jpeg artifacts, signature, watermark, username, signature, simple background, conjoined,</code>Prompting PonyFor all version I recommend you use: <code>score_9, score_8_up, score_7_up, score_6_up,</code> in your prompt and <code>score_4, score_5, censored, mutated, bad hands, deformed hands, missing fingers, multiple fingers, multiple digits, extra digits,</code> in your negatives. It will greatly improve the results.Version 1.0: This version has some mistakes in the naming of the lora. This is fixed in version 1.1, but this version has a shine/smoothing version 1.1 doesn't have.Version 1.1: I rebuild this lora with the same dataset from version 1.0 and now with the correct naming.If you need a place to start take a look at the showcase images above.

Hairu Style - Pony / Illustrious

hairu_style_v2_epoch_7.safetensors

Illustrious v1

Illustrious v2

v1.1 Pony

v1 Pony

The style of Hairu.
I made this lora to try and reproduce the style of Hairu.

The above showcase images were made with Addetailer extension enabled and using a face, eyes and hand model.

This LoRa is trained on Pony V6 and Illustrious 0.1, but I made the above showcase images in different checkpoints, because their style was more to my liking.

The Trigger is: ha1ru,

Advised strength: 0.6-0.9

Prompting Illustrious

For all versions I recommend you use: masterpiece, best quality, amazing quality, very aesthetic, high resolution, in your prompt and I myself am using the following negative prompt: text, abstract, glitch, deformed, mutated, ugly, disfigured, long body, lowres, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, lowres, (worst quality, bad quality:1.2), bad anatomy, jpeg artifacts, signature, watermark, username, signature, simple background, conjoined,

Files with SHA256: 97e7f558945cd6045e0d6690db0756829500eec7e48f7e8619f3515ee5bc4798