Hikari Noob v-pred 0.5
image/jpeg Fine-tuned NoobAI-XL(v-prediction) and merged SPO LoRA
NoobAI-XL(v-prediction)をファインチューンし、SPOをマージしました。
Features/特徴
Improved stability and quality.
Works with samplers other than Euler.
Good results with only 10 steps (12 steps or more recommended)
Fixed a problem in which the quality of output was significantly degraded when the number of tokens exceeded 76.
The base style is not strong and can be restyled by prompts or LoRAs.
安定性と品質を改善
わずか10ステップでよい結果を得られます(ただし12ステップ以上を推奨)
Zero Terminal SNRの代わりにNoise Offsetを使用することでEuler以外のサンプラーでも利用できるようにしました。
トークン数が76を超えると出力の品質が著しく低下する問題を修正しました。
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。
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の導入手順
Switch branch to dev (Run this command in the root directory of the webui: git checkout -b dev origin/dev or use Github Desktop)
Use the model as usual!
(日本語)
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)通常通りモデルを使用します。
Example Workflow for ComfyUI / ComfyUIサンプルワークフロー
Prompt Guidelines / プロンプト記法
Almost same as the base model/ベースモデルとおおむね同じ
To improve the quality of background, add simple background, transparent background to Negative Prompt.
Recommended Prompt / 推奨プロンプト
Positive: None/無し(Works good without masterpiece, best quality / masterpiece, best quality無しでおｋ)
Negative: worst quality, low quality, bad quality, lowres, jpeg artifacts, unfinished, photoshop \(medium\), abstract or empty(または無し)
Recommended Settings / 推奨設定
Steps: 10-24
Sampler: DPM++ 2M(dpmpp_2m)
Scheduler: Simple
Guidance Scale: 3.5-7
Hires.fix
Hires upscaler: 4x-UltraSharp or Latent(nearest-exact)
Denoising strength: 0.4-0.5(0.65-0.7 for latent)
Merge recipe(Weighted sum)
I made 6 Illustrious-based models and merged them.
Stage 0: finetunes v-pred test model with AI-generated images
Stage 1: finetunes stage 0 model with 300 scenery images from Gelbooru
Stage 2: Finetune and merge(see below)
*A-F,sd15: finetuned stage1(ReLoRA)
A * 0.6 + B * 0.4 = tmp1
tmp1 * 0.6 + C * 0.4 = tmp2
tmp2 * 0.7 + F * 0.3 = tmp3
tmp3 * 0.7 + E * 0.3 = tmp4
tmp4 * 0.5 + D * 0.5 = tmp5
tmp5 * 0.65 + sd15 * 0.35 = tmp6
tmp6 + SPO LoRA = Result
Training scripts:
Notice
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.
