Hikari Noob v-pred

CHECKPOINT
Reprint


Updated:

creator: RedRayz

NoobAI-XL(v-pred 1.0) based anime model

Hugging Face Repo(1.0.0): https://huggingface.co/RedRayz/hikari_noob_1.0.0

Hugging Face Repo(1.1.0): https://huggingface.co/RedRayz/hikari_noob_1.1.0

Hugging Face Repo(1.2.1): https://huggingface.co/RedRayz/hikari_noob_v-pred_1.2.1

Recommended Prompt/推奨プロンプト

Positive: None(works good without "masterpiece, best quality")

Negative: Empty(to get a best result, DO NOT USE NEGATIVE PROMPT. "worst quality", "old" and "lowres" have NO effects)

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"worst quality"、"old"および"lowres"の効果は無い)

About this checkpoint

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イラストを生成します。

This model does not have the "AI Masterpiece" style.

Features/特徴

  • Improved stability and quality.

  • The base style is not strong and can be restyled by prompts or LoRAs.

  • This model does not include any base model other than NoobAI (v-prediction), so it has the equivalent knowledge.

  • True v-prediction model with ZTSNR, No epsilon models merged

  • LoRAs for NoobAI-XL(v-prediction) work fine.

  • Works good without negative prompt/quality tags(v1.2.0 and newer)

  • 安定性と品質を改善

  • 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。

  • 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。

  • NoobAI-XL(v-prediction)用のLoRAがしっかり効きます。

  • 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。

  • Negative prompt/quality tag無しで良好な結果を得られます(v1.2.0以降)

Known issue(s)/既知の問題

  • Negative Prompt may degrade output quality.

  • Negative Promptを使用すると出力品質が低下する場合があります。

Requirements/動作要件

  • AUTOMATIC1111 WebUI on dev branch / devブランチ上のAUTOMATIC1111 WebUI

  • Latest version of ComfyUI / 最新版のComfyUI

  • Latest version of Forge or reForge / 最新版のForgeまたはreForge

Instruction for AUTOMATIC1111 / AUTOMATIC1111の手順

  1. Switch branch to dev(Run this command in the root directory of the webui: git checkout -b dev origin/dev or use Github Desktop)

  2. Use the model as usual!

(日本語)

  1. 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: git checkout -b dev origin/dev またはGithub Desktopを使う)

  2. 通常通りモデルを使用します。

Recommended settings/推奨設定

Steps: 14-24

Scheduler: Simple or SGM Uniform

Guidance Scale: 2-5(best value is 4)

Recommended Samplers/推奨サンプラー

  • DPM++ 2M

  • DPM++ 2M SDE

  • DPM++ 2M SDE Heun

  • Euler

Incompatible Samplers/非対応のサンプラー

  • Restart(Heun-based, when scond_order = True)

  • LMS

  • PLMS

  • Heun

  • LCM

  • Kohaku_LoNyu_Yog

other samplers are not tested/good.

Incompatible Schedulers/非対応のスケジューラー

  • Uniform

  • Karras

  • Beta

Recommended settings for Hires.fix/Hires.fixの推奨設定

Upscaler: 4x-UltraSharp(or ESRGAN-based model), Lanczos

Denoising strength: 0.4-0.6

Training information

Please visit the Hugging Face repository

Training script

sd-scripts

Recommended settings/推奨設定

Steps: 12-24

Scheduler: Simple or SGM Uniform

Guidance Scale: 2-5(best value is 4)

Note: "Align Your Steps" scheduler produces a very different image from other samplers.

Recommended Samplers/推奨サンプラー

  • DPM++ 2M (dpmpp_2m)

  • DPM++ 2M SDE

  • DPM++ 2M SDE Heun

  • Euler

  • Euler a (euler_ancestral)

Incompatible Samplers/非対応のサンプラー

  • Restart (Heun-based, when scond_order = True)

  • LMS (low quality)

  • PLMS (low quality)

  • Heun (same as restart)

  • LCM (NaN error)

  • Kohaku_LoNyu_Yog

  • UniPC (very unstable, low quality)

other samplers are not tested/good.

Incompatible Schedulers/非対応のスケジューラー

  • Uniform (makes artifacts like glaze or nightshade)

  • Karras (unstable)

  • Beta (makes blurry images)

Recommended settings for Hires.fix/Hires.fixの推奨設定

Upscaler: 4x-UltraSharp(or ESRGAN-based model), Lanczos, Latent(nearest-exact)

Denoising strength: 0.4-0.6 (0.65-0.7 for latent)

Training information

Please visit the Hugging Face repository

Training script

sd-scripts

Notice/おしらせ:

This model is licensed under Fair AI Public License 1.0-SD

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Fair AI Public License 1.0-SD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。

Recommended Prompt/推奨プロンプト

To improve the quality of background, add "simple background, transparent background" to Negative Prompt.

背景の品質を向上させるには、"simple background, transparent background"をネガティブプロンプトに追加してください。

Positive: None(works good without "masterpiece, best quality")

Negative: "worst quality, low quality, bad quality, lowres, jpeg artifacts, unfinished, photoshop \(medium\), abstract", or empty

Anime Style

Positive: "anime screenshot, anime screencap, anime coloring"

Negative: "worst quality, oldest"

Blue Archive Style

Positive: "official art, blue archive, game cg, non-web source, newest"

Negative: "worst quality, bad quality, low quality, lowres, photoshop \(medium\), abstract, oldest"

Recommended settings for DMD2 LoRA

LoRA strength: 0.4

Steps: 6-10

Sampler: Euler/Euler a

Scheduler: Simple

Guidance Scale: 1-1.5

Note: DMD2 will change the style of the model.

The model already exists. If you are the creator, please contact us to claim it.
View duplicate model

Version Detail

Illustrious

Project Permissions

Model reprinted from : https://civitai.com/models/938672

Reprinted models are for communication and learning purposes only, not for commercial use. Original authors can contact us to transfer the models through our Discord channel --- #claim-models.

Related Posts