This is Arctenox's (my) first refiner LoRA that has 1k+ images. You may use it if you want. Although you'll have to test strengths and what not yourself lol. Pretty self explanatory. It should make models be more basic unless you prompt for semirealism, etc.<h1 id="v1.0-illustrious:-for-any-illustrious-model.-the-one-you-want-to-use-if-you're-not-using-ionsyx-or-my-illunoob-mix-n9v8tift5">V1.0 - Illustrious: For any illustrious model. The one you want to use if you're not using Ionsyx or my IlluNoob Mix</h1><h1 id="v1.0-ionsyx:-for-the-ionsyx-checkpoint-model.-using-another-model-will-not-get-you-the-same-results-as-seen-in-the-thumbnail.-w1lseqbpr">V1.0 - Ionsyx: For the Ionsyx Checkpoint Model. Using another model will not get you the same results as seen in the thumbnail.</h1><h1 id="v1.0-illunoob-mix:-will-be-removed-as-it's-not-being-used-on-the-model-it's-intended-for.-k8vekjypy">V1.0 - IlluNoob Mix: Will be Removed as it's not being used on the model it's intended for.</h1><h3 id="note:-lora-is-intended-so-you-don't-need-to-mess-with-weights.-if-you-need-to-mess-with-weights-please-use-the-illustrious-version-if-it's-an-illustrious-model.-xw54fti0p">Note: LoRA is intended so you don't need to mess with weights. If you need to mess with weights please use the Illustrious version if it's an illustrious model.</h3><h3 id="disclaimer:-this-is-not-a-style-lora-so-don't-use-it-as-one-and-expect-a-style-from-it.-and-it-may-not-work-if-you-don't-use-the-respective-models-mentioned-excluding-the-illustrious-one-which-is-for-illustrious-models-in-general.-17ze6majd">Disclaimer: This is not a style LoRA so don't use it as one and expect a style from it. And it may not work if you don't use the respective models mentioned, excluding the Illustrious one which is for Illustrious models in general.</h3> Made ones for Illustrious, Arctenox's (my) IlluNoob Mix and Ionsyx V3.5. Note: Spent 3640 buzz on all three version V1.0's. Also no trigger words are needed. :D Also when each version of this gets 1,000 Images ill update them to the next version. Otherwise they'll stay as V1.0.⚠️ IF YOU'RE GOING TO STEAL MY (<a target="_blank" rel="ugc" href="https://civarchive.com/user/Arctenox">https://civarchive.com/user/Arctenox</a>) MODELS OR POST THEM SOMEWHERE ELSE, MAKE IT SO THE DOWNLOAD IS ENABLED, IF IT'S NOT THEN I KNOW A BOT UPLOADED IT.⚠️ <a target="_blank" rel="ugc" href="https://civarchive.com/models/1580340">https://civarchive.com/models/1580340</a>

Arctenox's Refiner LoRAs

This version was trained on my Ionsyx V3.5, not intended for other models. Same dataset and LoRA training settings, as the illunoob mix version, excluding clip skip which is now 2.<a target="_blank" rel="ugc" href="https://civitai.com/models/1580340?modelVersionId=1791406">https://civitai.com/models/1580340?modelVersionId=1791406</a>

Files with SHA256: 3d7bdf6a6cfc71a048372f185d5172a7950c6b8ffe74be1c222c85f1a78830a2