NoobAI-XL(v-pred 1.0) based anime modelHugging Face Repo(1.2.4): <a target="_blank" rel="ugc" href="https://huggingface.co/RedRayz/hikari_noob_v-pred_1.2.4">https://huggingface.co/RedRayz/hikari_noob_v-pred_1.2.4</a><h2 id="recommended-prompt">Recommended Prompt/推奨プロンプト</h2>Positive: None(works good without "masterpiece, best quality")Negative: Empty(to get a best result, DO NOT USE NEGATIVE PROMPT(use NegPiP instead). negative quality tags have NO effect)ポジティブ: なし(品質タグなしでok)ネガティブ: 空(最良の結果を得るにはnegative promptを使用しないこと。低品質な成分は除去済みのためネガティブ品質タグの効果は無い)プロンプトの注意: 一般的なepsilonモデルと比較して、プロンプトに書いていないものは一切生成しない性質が強いため、短すぎるプロンプトでは相応に地味、あるいは多様だがやや不安定な結果になります。良い結果を得るために長いプロンプトを推奨します。思いつかないならTIPO(DanTagGen)を使うといいよ！背景をより豪華にしたい場合はNegPiPでsimple backgroundを負の強度で使用してください。<h2 id="about-this-checkpoint">About this checkpoint</h2>Fine-tuned NoobAI-XL(v-prediction) without merging another base models, and merged SPO(Step-aware Preference Optimization) LoRA他所のベースモデルを混ぜずに調整し、SPO LoRAをマージしたNoobAI-XL(v-prediction)NoobAI-XL(v-prediction版)に特定画風を学習させたcheckpointを複数作成しマージしたものです。AIマスピ絵のくどい感じやわざとらしい背景が好みではないので作りました。AI特有のくどい感じを抑えつつ鮮やかで美しい、かつ多様なイラストを生成します。なお、AIらしさを抑えたため、マージモデルのようなはっきり・くっきりした画像は生成しない傾向があります。Note : This model does not have the "AI Masterpiece" style and any default style.注意: このモデルにはAIマスピ画風とデフォルトの画風はありません。<h2></h2><h2 id="features">Features/特徴</h2><ul><li>Improved stability and quality.</li><li>The base style is not strong and can be restyled by prompts or LoRAs.</li><li>This model does not include any base model other than NoobAI (v-prediction), so it has the equivalent knowledge.</li><li>True v-prediction model with ZTSNR, No epsilon models merged</li><li>LoRAs for NoobAI-XL(v-prediction) work fine.</li><li>Works good without negative quality tags(v1.2.0 and newer)</li><li>This excels in compositional diversity.</li><li>安定性と品質を改善</li><li>素の画風は強くないので、プロンプトやLoRAによる画風変更ができます。</li><li>このモデルはNoobAI(v-prediction)以外のベースモデルを一切含まず、それと同等の知識があります。</li><li>NoobAI-XL(v-prediction)用のLoRAがしっかり効きます。</li><li>真のv-prediction + ZTSNRなモデルです。Epsilonモデルは一切マージしていません。</li><li>ネガティブ品質タグ無しで良好な結果を得られます(v1.2.0以降)</li><li>優れた構図の多様性</li></ul><h3 id="known-issue(s)">Known issue(s)/既知の問題</h3><ul><li>Negative Prompt may degrade output quality. Please use NegPiP instead.</li><li>Negative Promptを使用すると出力品質が低下する場合があります。代わりにNegPiPを使用してください。</li></ul><h2 id="requirements">Requirements/動作要件</h2><ul><li>AUTOMATIC1111 WebUI on <code>dev</code> branch / devブランチ上のAUTOMATIC1111 WebUI</li><li>Latest version of ComfyUI / 最新版のComfyUI</li><li>Latest version of Forge or reForge / 最新版のForgeまたはreForge</li></ul><h3 id="instruction-for-automatic1111-automatic1111">Instruction for AUTOMATIC1111 / AUTOMATIC1111の手順</h3><ol><li>Switch branch to <code>dev</code>(Run this command in the root directory of the webui: <code>git checkout -b dev origin/dev</code> or use Github Desktop)</li><li>Use the model as usual!</li></ol>Tip: reForge is better than 1111 webui. The original webui has been abandoned.(日本語)<ol><li>devブランチに切り替えます(次のコマンドをwebui直下で実行します: <code>git checkout -b dev origin/dev</code> またはGithub Desktopを使う)</li><li>通常通りモデルを使用します。</li></ol><h3></h3><h3 id="recommended-settings">Recommended settings/推奨設定</h3>Steps: 14-24Scheduler: Simple, Turbo, SGM UniformGuidance(CFG) Scale: 2-5(best value is 3.5)<h3 id="recommended-samplers">Recommended Samplers/推奨サンプラー</h3><ul><li>DPM++ 2M</li><li>DPM++ 2M SDE</li><li>DPM++ 2M SDE Heun</li><li>Euler</li></ul>Note: SDE(ancestral) samplers may reduce contrast.<h3 id="incompatible-samplers">Incompatible Samplers/非対応のサンプラー</h3><ul><li>Restart(Heun-based)</li><li>LMS</li><li>PLMS</li><li>Heun</li><li>LCM</li><li>Kohaku_LoNyu_Yog</li></ul>other samplers are not tested/good.<h3 id="incompatible-schedulers">Incompatible Schedulers/非対応のスケジューラー</h3><ul><li>React Consinusodial DynSF</li><li>Phi</li><li>Laplace</li><li>Align Your Steps</li></ul><h3 id="recommended-settings-for-hires.fixhires.fix">Recommended settings for Hires.fix/Hires.fixの推奨設定</h3>Upscaler: 4x-UltraSharp(or ESRGAN-based model), Lanczos or LatentDenoising strength: 0.4-0.6(&gt;0.55 for Latent)<h2 id="training-information">Training information</h2>Please visit the Hugging Face repository<h2 id="training-script">Training script</h2><a target="_blank" rel="ugc" href="https://github.com/kohya-ss/sd-scripts">sd-scripts</a><h2 id="notice:">Notice/おしらせ:</h2>This model is licensed under <a target="_blank" rel="ugc" href="https://freedevproject.org/faipl-1.0-sd/">Fair AI Public License 1.0-SD</a>If you make modify this model, you must share both your changes and the original license.You are prohibited from monetizing any close-sourced fine-tuned / merged model, which disallows the public from accessing the model's source code / weights and its usages.このモデルは<a target="_blank" rel="ugc" href="https://freedevproject.org/faipl-1.0-sd/">Fair AI Public License 1.0-SD</a>でライセンスされています。このモデルを改変する場合、変更点および元のライセンスを明記する必要があります。モデルのソースコード/ウェイトや使用法などを閲覧できないクローズソースのファインチューン/マージモデルの収益化は禁止されています。

Hikari Noob v-pred

The config file is required for AUTOMATIC1111 on "main" branch.AUTOMATIC1111 on "dev", ComfyUI and Forge/reForge will work without it.<ul><li>v0.5 was too flat, so we improved the quality of detail rendering and made outline thinner!</li><li>Improved the quality of the landscapes just a little bit</li><li>Reduced skin color overshoot?</li><li>v0.5は絵柄が平坦すぎたのでディティールを増やしてアウトラインを細くしました。</li><li>ほんの少しだけ風景画の品質が改善</li><li>肌の色のオーバーシュート(白飛び)を軽減?</li></ul>

hikariNoobVPred_06.yaml

hikariNoobVPred_06.safetensors

1.2.4

1.2.3

1.2.2

1.2.1

1.2.0

1.1.0

1.0.1

1.0.0 0.7.5s

0.6.5s

0.6.5 0.6TurboTest

Calico Cat Tower is <a target="_blank" rel="ugc" href="https://civarchive.com/models/950531">RouWei</a> based checkpoint.VAE is baked in.RouWei has its own specific features and prompting, so please refer to the <a target="_blank" rel="ugc" href="https://civarchive.com/models/950531">RouWei</a> page for instructions on how to use it.You need to use the webui that support v-prediction if you use v-prediction version.<ul><li>ComfyUI (recommended)</li><li>Forge Classic (recommended)</li><li>reForge</li><li>Forge</li><li>AUTOMATIC1111 (place the config file in the same folder as the checkpoint)</li></ul><h3 id="license-yg6rfqo7j">License</h3>This model merges the NoobAI-Xl model, so the license follows <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294">NoobAI-XL License (Modified Fair AI Public License 1.0-SD)</a> and is available for non-commercial use only. This license prohibits any form of commercialization, including but not limited to monetization or commercial use of the model, derivative models, or model-generated products.<h3 id="recommended-setting-for-v-prediction-version.-8rxs8s8oo">Recommended Setting for V-prediction version.</h3><ul><li>Steps: 20-30</li><li>CFG scale: 3-5</li><li>Sampler: Euler a</li><li>Use ADetailer as needed</li></ul><h3 id="recommended-setting-for-epsilon-prediction-version.-6xwbrki00">Recommended Setting for Epsilon-prediction version.</h3><ul><li>Steps: 20-30</li><li>CFG scale: 3-7</li><li>Sampler: Euler a, DPM++ 2M Karras</li><li>Use ADetailer as needed</li></ul><h3 id="positive-prompt-ucuu7dloi">Positive Prompt</h3><pre><code>masterpiece, best quality</code></pre>Enter the quality tag and artist tag as shown in the example below, otherwise the image quality will be reduced.<ul><li>Add 'BREAK' after them (for A1111 and A1111 derivatives such as Forge, recommended)</li><li>Use conditioning concat node (for ComfyUI, recommended)</li><li>Put them in the very end of prompt</li></ul><h3 id="negative-prompt-j8guvivym">Negative Prompt</h3><pre><code>worst quality, bad quality, low quality, lowres, scan artifacts, jpeg artifacts, sketch, light particles, watermark,</code></pre><h3 id="merge-recipes-(v-pred-v2.0)-0clx2sr20">Merge recipes (v-pred v2.0)</h3><ol><li>Merging the difference between <a target="_blank" rel="ugc" href="https://civarchive.com/models/950531?modelVersionId=1882934">Rouwei 0.80 vpred</a> and <a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0">stable-diffusion-xl-base-1.0</a>, <a target="_blank" rel="ugc" href="https://civarchive.com/models/312530?modelVersionId=1720768">CyberRealistic XL v5.6</a> and stable-diffusion-xl-base-1.0 to Rouwei 0.80 vpred by "Perpendicular Component" and "Add Difference" (Checkpoint A)</li><li>Merging Checkpoint A to Rouwei 0.80 vpred by "SLERP", alpha = 0.5 (Checkpoint B)</li><li>Merging Checkpoint B to Rouwei 0.80 vpred by "Rotate", alignment = 1.0, alpha = 0.0 (Checkpoint C)</li><li>Replacing the CLIP of Checkpoint C by Rouwei 0.80 vpred (Checkpoint D)</li><li>Finetuning Checkpoint D with anime style dataset(0.5k) by OFT method, dim = 4, alpha = 1e-3, learning rate = 2.5e-6, 15000 steps. And merging the OFT model to Checkpoint D, ratio = 1.0 (Checkpoint E)</li><li>Merging Checkpoint E to Checkpoint D by "Weighted Sum" with block (0,0,0,0,0,0,0,0,0,0,0,1,1,0,0,0,0,0,0,0) (Checkpoint F)</li><li>Merging the <a target="_blank" rel="ugc" href="https://civarchive.com/models/920709?modelVersionId=1495826">Cat Tower vpred v1.7</a> to Checkpoint F by "SLERP" with block (0,0,0,0,0,0,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0,0,0,0,0) (Checkpoint G)</li><li>Merging the difference between <a target="_blank" rel="ugc" href="https://civarchive.com/models/1277670?modelVersionId=1772645">JANKU v4.0</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/950531?modelVersionId=1330983">RouWei v0.7 epred</a> to Checkpoint G by "Train Difference" with block (0,0,0,0,0,0,0,0,0,0,0,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint H)</li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/1027203/noobvpencil-xl">Add v_pred and ztsnr keys to Checkpoint H by noob_v_pencil-XL merge recipes</a> (Calico Cat Tower v-pred v2.0)</li></ol><h3 id="merge-recipes-(e-pred-v2.0)-3oocoa656">Merge recipes (e-pred v2.0)</h3><ol><li>Merging the difference between <a target="_blank" rel="ugc" href="https://huggingface.co/Minthy/RouWei-0.8">rouwei_080_epsilon_fp16</a> and <a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0">stable-diffusion-xl-base-1.0</a>, <a target="_blank" rel="ugc" href="https://civarchive.com/models/312530?modelVersionId=1720768">CyberRealistic XL v5.6</a> and stable-diffusion-xl-base-1.0 to rouwei_080_epsilon_fp16 by "Perpendicular Component" and "Add Difference" (Checkpoint A)</li><li>Merging Checkpoint A to rouwei_080_epsilon_fp16 by "SLERP", alpha = 0.5 (Checkpoint B)</li><li>Merging Checkpoint B to rouwei_080_epsilon_fp16 by "Rotate", alignment = 1.0, alpha = 0.0 (Checkpoint C)</li><li>Replacing the CLIP of Checkpoint C by rouwei_080_epsilon_fp16 (Checkpoint D)</li><li>Finetuning Checkpoint D with anime style dataset(0.5k) by OFT method, dim = 4, alpha = 1e-3, learning rate = 2.5e-6, 15000 steps. And merging the OFT model to Checkpoint D, ratio = 1.0 (Checkpoint E)</li><li>Merging Checkpoint E to Checkpoint D by "Weighted Sum" with block (0,0,0,0,0,0,0,0,0,0,0,1,1,0,0,0,0,0,0,0) (Checkpoint F)</li><li>Merging the <a target="_blank" rel="ugc" href="https://civarchive.com/models/920709?modelVersionId=1776050">Cat Tower epred v1.3</a> to Checkpoint F by "SLERP" with block (0,0,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0,0,0,0,0) (Checkpoint G)</li><li>Merging the difference between <a target="_blank" rel="ugc" href="https://civarchive.com/models/1277670?modelVersionId=1772645">JANKU v4.0</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/950531?modelVersionId=1330983">RouWei v0.7 epred</a> to Checkpoint G by "Train Difference" with block (0,0,0,0,0,0,0,0,0,0,0,0.2,0.2,0.2,0.2,0,0,0,0,0) (Calico Cat Tower e-pred v2.0)</li></ol><h3 id="merge-recipes-(v-pred-v1.0)-bzax7q4fs">Merge recipes (v-pred v1.0)</h3><ol><li>Finetuning <a target="_blank" rel="ugc" href="https://civarchive.com/models/950531?modelVersionId=1360303">RouWei v0.7 vpred</a> with anime style dataset(0.5k) by OFT method, dim = 8, alpha = 1e-3, learning rate = 1e-4, 15000 steps</li><li>Merging the OFT model to <a target="_blank" rel="ugc" href="https://civarchive.com/models/950531?modelVersionId=1360303">RouWei v0.7 vpred</a>, ratio = 1.0 (Checkpoint A)</li><li>Merging Checkpoiint A to <a target="_blank" rel="ugc" href="https://civarchive.com/models/950531?modelVersionId=1360303">RouWei v0.7 vpred</a> by "Weighted Sum", alpha = 0.3 (Checkpoint B)</li><li>Merging the difference between <a target="_blank" rel="ugc" href="https://civarchive.com/models/860278?modelVersionId=1416446">Cat Carrier v3.0</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/1232765?modelVersionId=1389133">Illustrious XL v1.0</a> to Checkpoint B by "Add Difference" with block (0,0,0,0,0,0,0,0,0,0,0,0.5,0.5,0.5,0.5,0,0,0,0,0) (Checkpoint C)</li><li>Merging the difference between <a target="_blank" rel="ugc" href="https://civarchive.com/models/920709?modelVersionId=1443338">Cat Tower vpred v1.6</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1190596">NoobAI-V-Pred-1.0-Version</a> to Checkpoint C by "Train Difference" with block (0,0,0,0,0,0,0,0,0,0,0,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint D)</li><li>Merging the difference between <a target="_blank" rel="ugc" href="https://civarchive.com/models/615558">Catloaf</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/257749/pony-diffusion-v6-xl?modelVersionId=290640">ponyDiffusionV6XL</a> to Checkpoint D by "Train Difference" with block (0,0,0,0,0,0,0,0,0,0,0,0.15,0.15,0.15,0.15,0,0,0,0,0) (Checkpoint E)</li><li>Merging the difference between <a target="_blank" rel="ugc" href="https://civarchive.com/models/894218?modelVersionId=1222866">copycat-noob Vpred_v1.01</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1190596">NoobAI-V-Pred-1.0-Version</a> to Checkpoint E by "Train Difference" with block (0,0,0,0,0,0,0,0,0,0,0,0.15,0.15,0.15,0.15,0,0,0,0,0) (Checkpoint F)</li><li>Merging style LoRA to Checkpoint F (Checkpoint G)</li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/1027203/noobvpencil-xl">Add v_pred and ztsnr keys to Checkpoint G by noob_v_pencil-XL merge recipes</a> (Calico Cat Tower v-pred v1.0)</li></ol>

Calico Cat Tower

<ul><li>RouWei v0.8.0 vpred based</li></ul>

calicoCatTower_v20VPred.yaml

calicoCatTower_v20VPred.safetensors

v2.0 (v-pred)

v2.0 (e-pred)

v1.0 (vpred)

Kitten Tower is NoobAI XL model focused on cute style.VAE is baked in.You need to use the webui that support v-prediction.<ul><li>ComfyUI</li><li>reForge</li><li>AUTOMATIC1111 (place the config file in the same folder as the checkpoint)</li></ul><h3 id="license-w7h1azqa9">License</h3>This model adheres to the <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294">NoobAI-XL License (Modified Fair AI Public License 1.0-SD)</a> and is available for non-commercial use only. This license prohibits any form of commercialization, including but not limited to monetization or commercial use of the model, derivative models, or model-generated products.<h3 id="recommended-setting-g0jgs9mkv">Recommended Setting</h3><ul><li>Steps: 25-30</li><li>CFG scale: 5-7</li><li>Sampler :Euler a, EulerI recommend using "Euler" if you use v-pred model. Using "Euler a" sometimes generates colored skin and objects such as red, blue and so on.</li><li>Scheduler for vPred : Normal, Simple (Other schedulers produce corrupt images)</li><li>Use ADetailer as needed</li></ul><h3 id="positive-prompt-tzdzbr5ve">Positive Prompt</h3><pre><code>masterpiece, best quality</code></pre><h3 id="negative-prompt-p2wow5zxp">Negative Prompt</h3><pre><code>worst quality, bad quality, low quality, lowres, scan artifacts, jpeg artifacts, sketch, light particles</code></pre><h3 id="merge-recipes-(epsilon-pred-v1.0)-0da6n8b92">Merge recipes (Epsilon-pred v1.0)</h3><ol><li>Merging the difference between <a target="_blank" rel="ugc" href="https://civarchive.com/models/312530?modelVersionId=1457203">CyberRealistic XL v5</a> and <a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0">stable-diffusion-xl-base-1.0</a> to <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1116447">NoobAI-XL Epsilon-pred v1.1</a> by "Add Difference" (Checkpoint A)</li><li>Replacing the CLIP of Checkpoint A by <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1116447">NoobAI-XL Epsilon-pred v1.1</a> (Checkpoint B)</li><li>Merging <a target="_blank" rel="ugc" href="https://civarchive.com/models/860278?modelVersionId=1416446">Cat Carrier v3.0</a> to Checkpoint B by "Weighted Sum" with block (0,0,0,0,0,0,0,0,0.3,0.3,0.3,0,0.3,0.3,0.3,0,0,0.3,0,0) (Checkpoint C)</li><li>Merging <a target="_blank" rel="ugc" href="https://civarchive.com/models/920709?modelVersionId=1431383">Cat Tower Epsilon-pred v1.2</a> to Checkpoint C by "Add Cosine B" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint D)</li><li>Merge the difference between <a target="_blank" rel="ugc" href="https://huggingface.co/da2el/ioliPonyMix">ioliPonyMix v3.1</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/257749/pony-diffusion-v6-xl?modelVersionId=290640">ponyDiffusionV6XL</a> to Checkpoint D by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint E)</li><li>Merge the difference between <a target="_blank" rel="ugc" href="https://civarchive.com/models/935880?modelVersionId=1248230">IllusionBreed v3.0</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1116447">NoobAI-XL-Epsilon-pred 1.1-Version</a> to Checkpoint E by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0,0,0,0) (Checkpoint F)</li><li>Merge the difference between CatPod v1.0(I merged it but not published) and <a target="_blank" rel="ugc" href="https://civarchive.com/models/257749/pony-diffusion-v6-xl?modelVersionId=290640">ponyDiffusionV6XL</a> to Checkpoint F by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint G)</li><li>Merge style LoRA to Checkpoint G (Kitten Tower Epsilon-pred v1.0)</li></ol><h3 id="merge-recipes-(vpred-v2.0)-r6n22sjvl">Merge recipes (vPred v2.0)</h3><ol><li>Merging the difference between <a target="_blank" rel="ugc" href="https://civarchive.com/models/312530?modelVersionId=1457203">CyberRealistic XL v5</a> and <a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0">stable-diffusion-xl-base-1.0</a> to <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1190596">NoobAI-V-Pred-1.0-Version</a> by "Add Difference" (Checkpoint A)</li><li>Replacing the CLIP of Checkpoint A by <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1190596">NoobAI-V-Pred-1.0-Version</a> (Checkpoint B)</li><li>Merging <a target="_blank" rel="ugc" href="https://civarchive.com/models/860278?modelVersionId=1416446">Cat Carrier v3.0</a> to Checkpoint B by "Weighted Sum" with block (0,0,0,0,0,0,0,0,0.4,0.4,0.4,0,0.4,0.4,0.4,0,0,0.4,0,0) (Checkpoint C)</li><li>Merging <a target="_blank" rel="ugc" href="https://civarchive.com/models/1195325?modelVersionId=1345854">Kitten Tower vpred v1.0</a> to Checkpoint C by "Add Cosine B" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint D)</li><li>Merge the difference between <a target="_blank" rel="ugc" href="https://huggingface.co/da2el/ioliPonyMix">ioliPonyMix v3.1</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/257749/pony-diffusion-v6-xl?modelVersionId=290640">ponyDiffusionV6XL</a> to Checkpoint D by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint E)</li><li>Merge the difference between <a target="_blank" rel="ugc" href="https://civarchive.com/models/935880?modelVersionId=1248230">IllusionBreed v3.0</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1116447">NoobAI-XL-Epsilon-pred 1.1-Version</a> to Checkpoint E by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0,0,0,0) (Checkpoint F)</li><li>Merge the difference between CatPod v1.0(I merged it but not published) and <a target="_blank" rel="ugc" href="https://civarchive.com/models/257749/pony-diffusion-v6-xl?modelVersionId=290640">ponyDiffusionV6XL</a> to Checkpoint F by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint G)</li><li>Merge style LoRA to Checkpoint G (Checkpoint H)</li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/1027203/noobvpencil-xl">Add v_pred and ztsnr keys to Checkpoint H by noob_v_pencil-XL merge recipes</a> (Kitten Tower vPred v1.0)</li></ol><h3 id="merge-recipes-(vpred-v1.0)-laf9o1tmk">Merge recipes (vPred v1.0)</h3><ol><li>Merge <a target="_blank" rel="ugc" href="https://civarchive.com/models/920709?modelVersionId=1200764">Cat Tower v1.5 vPred</a> to <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1190596">NoobAI-V-Pred-1.0-Version</a> by "Weighted Sum" with block (0,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0,0,0,0,0) (Checkpoint A)</li><li>Merge the difference between <a target="_blank" rel="ugc" href="https://huggingface.co/da2el/ioliPonyMix">ioliPonyMix v3.1</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/257749/pony-diffusion-v6-xl?modelVersionId=290640">ponyDiffusionV6XL</a> to Checkpoint A by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint B)</li><li>Merge the difference between <a target="_blank" rel="ugc" href="https://civarchive.com/models/935880?modelVersionId=1248230">IllusionBreed v3.0</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1116447">NoobAI-XL-Epsilon-pred 1.1-Version</a> to Checkpoint B by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0,0,0,0) (Checkpoint C)</li><li>Merge the difference between CatPod v1.0(I merged it but not published) and <a target="_blank" rel="ugc" href="https://civarchive.com/models/257749/pony-diffusion-v6-xl?modelVersionId=290640">ponyDiffusionV6XL</a> to Checkpoint C by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint D)</li><li>Merge style LoRA to Checkpoint D (Checkpoint E)</li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/1027203/noobvpencil-xl">Add v_pred and ztsnr keys to Checkpoint E by noob_v_pencil-XL merge recipes</a> (Kitten Tower vPred v1.0)</li></ol>

Kitten Tower

<ul><li>Adjusting the style</li><li>Improved generation of pure single color backgrounds</li></ul>

kittenTower_v20VPred.yaml

kittenTower_v20VPred.safetensors

v1.0 (eps-pred)

v1.0 (v-pred)

Cat Tower (NoobAI XL checkpoint)

<ul><li>NoobAI-V-Pred-0.75S-Version based</li><li>Adjusted the art style to be closer to v1.1 (epsilon pred)</li><li>Images can now be generated with Automatic1111</li></ul>

catTowerNoobaiXL_v14Vpred.safetensors

catTowerNoobaiXL_v14Vpred.yaml

v1.4 (vPred)

ChenkinNoob v1.1

ChenkinNoob v1.0

Obsession v2.0 (vPred)

v1.4 (Epsilon-pred)

v1.8 (vPred)

v1.3 (Epsilon-pred)

v1.7 (vPred)

v1.6 (vPred)

v1.2 (Epsilon-pred)

v1.5 (vPred)

v1.1 (Epsilon-pred)

v1.3 (vPred)

v1.2 (vPred)

v1.1 (vPred)

v1.0 (vPred)

v1.0 (Epsilon-pred)

<ul><li>NoobAI-V-Pred-1.0-Version based</li><li>Changed some of the models to be merged</li><li>v1.6 is merged Illustrious XL v1.0 based model, but v1.6 has less ability to generate high resolution images directly like Illustrious XL v1.0 </li></ul>

catTowerNoobaiXL_v16Vpred.safetensors

catTowerNoobaiXL_v16Vpred.yaml

<ul><li>NoobAI-V-Pred-1.0-Version based</li><li>The art style has been changed (like sharp and clear).</li></ul>

catTowerNoobaiXL_v18Vpred.safetensors

catTowerNoobaiXL_v18Vpred.yaml

<ul><li>RouWei v0.7 vpred based</li></ul>

calicoCatTower_v10Vpred.safetensors

calicoCatTower_v10Vpred.yaml

<ul><li>NoobAI-V-Pred-1.0-Version based</li><li>I canged some of the checkpoints to merge, but the merge recipes are almost the same as v1.4</li></ul>

catTowerNoobaiXL_v15Vpred.safetensors

catTowerNoobaiXL_v15Vpred.yaml

Cat Tower (NoobAI XL Checkpoint)

kittenTower_v10VPred.safetensors

kittenTower_v10VPred.yaml

<ul><li>Based on Obsession (Illustrious-XL) v-pred v2.0, which knows more characters than NoobAI XL</li><li>Changed the part of the merge recipe from merging OFT model to full finetune</li></ul>

catTowerNoobaiXL_obsessionV20Vpred.yaml

catTowerNoobaiXL_obsessionV20Vpred.safetensors

<ul><li>NoobAI-V-Pred-0.65S-Version based</li><li>It has become slightly easier to create dark images.</li></ul>

catTowerNoobaiXL_v12Vpred.yaml

catTowerNoobaiXL_v12Vpred.safetensors

<ul><li>NoobAI-V-Pred-0.6-Version based</li><li>Adjusted the art style to be closer to v1(eps version)</li><li>ZTSNR supported (maybe)</li></ul>

catTowerNoobaiXL_v11Vpred.safetensors

catTowerNoobaiXL_v11Vpred.yaml

<ul><li>NoobAI-V-Pred-1.0-Version based</li><li>Fixed an issue where v1.6 vPred would sometimes generate images with corrupt coloring when using certain prompts.</li><li>The art style has been changed slightly.</li></ul>

catTowerNoobaiXL_v17Vpred.yaml

catTowerNoobaiXL_v17Vpred.safetensors

<ul><li>NoobAI-V-Pred-0.65S-Version based</li><li>I changed the merge recipes because the previous merge recipes made too high brightness model.</li></ul>

catTowerNoobaiXL_v13Vpred.safetensors

catTowerNoobaiXL_v13Vpred.yaml

This checkpoint is v-prediction version.You need to use the webui that support v-prediction.

Files with SHA256: e1bffa820ca38f9f61fd9d881ccfb76df9c9ca762483101ed2212151032d3f70