motimalu/kirazuri-lazuli-noobai-xl-vpred-2.0

Kirazuri Lazuli (Noobai V-Pred) 2.0

Model Introduction

This checkpoint is a personal project trained locally from NoobAI-XL (NAI-XL) V-Pred 1.0-Version on a 4090 with a small dataset of 14,065 total images.

It focuses on adding additional knowledge since the data cutoff of the base model (2024/10/24), including styles, concepts, and characters from anime, video games, and virtual youtubers.

Model Description

• Developed by: [motimalu]
• Model Type: Diffusion-based text-to-image generative model
• Finetuned from model NoobAI XL V-Pred 1.0

How to Use the Model

Your preferred generation settings used for NoobAI-XL (NAI-XL) V-Pred 1.0-Version should be mostly transferrable.

Refer to https://huggingface.co/Laxhar/noobai-XL-Vpred-1.0#how-to-use-the-model

Previews are generated with a ComfyUI workflow using DynamicThresholdingFull, Upscaling, and FaceDetailer, metadata is contained in the images.

DynamicThresholding (CFG-Fix) settings used with a CFG of 10:

dynthres_enabled: True, dynthres_mimic_scale: 7, dynthres_threshold_percentile: 1, dynthres_mimic_mode: Half Cosine Down, dynthres_mimic_scale_min: 1, dynthres_cfg_mode: Half Cosine Down, dynthres_cfg_scale_min: 3, dynthres_sched_val: 1, dynthres_separate_feature_channels: enable, dynthres_scaling_startpoint: ZERO, dynthres_variability_measure: STD, dynthres_interpolate_phi: 1
For samplers, recommendation is Euler for generation, Euler Ancestral for upscaling/inpainting.

Recommended prompt structure:

Quality modifiers masterpiece, best quality, very aesthetic should be positioned at the end of the prompt.

Artist names can be prefixed with artist: to prevent token bleeding with artist names and concepts.

A1111 schedule prompting syntax is used in ComfyUI through the comfyui-prompt-control extension to combine artist styles, i.e: artist:[artist1|artist2|artist3]

Positive prompt:

{{characters}}, {{copywrites}}, {{artists}},
{{tags}},
absurdres, masterpiece, best quality, very aesthetic

Training details

The kohya-ss/sd-scripts training configs used can be found on github.

Version 2.0 now has a much better representation of all the characters, concepts, and styles I hoped to train for this checkpoint.

Single training run on a full dataset, expanded with more recent data:

• Dataset cutoff: 2025/06/13
• Training images: 14,065
• Regularization images: 7056 (Generated from NoobAI-XL (NAI-XL) V-Pred 1.0-Version)
• Optimizer: Adafactor
• Training precision: Full-fat fp32
• Batch size: 4
• U-Net LR: 6e-6
• TE LR: 2e-6
• Epochs: 50
• Steps: 352,600 (~344 GPU hours at 3.52s/it)

Recognitions

Thanks to Laxhar Lab for the NoobAI-XL (NAI-XL) V-Pred 1.0-Version base model.

Thanks to narugo1992 and the deepghs team for open-sourcing various training sets, image processing tools, and models.

Thanks to kohya-ss for the sd-scripts trainer.

Model License

This model's license inherits from https://huggingface.co/OnomaAIResearch/Illustrious-xl-early-release-v0 fair-ai-public-license-1.0-sd and adds the following terms. Any use of this model and its variants is bound by this license.

I. Usage Restrictions

• Prohibited use for harmful, malicious, or illegal activities, including but not limited to harassment, threats, and spreading misinformation.
• Prohibited generation of unethical or offensive content.
• Prohibited violation of laws and regulations in the user's jurisdiction.

II. Commercial Prohibition

We prohibit any form of commercialization, including but not limited to monetization or commercial use of the model, derivative models, or model-generated products.

III. Open Source Community

To foster a thriving open-source community,users MUST comply with the following requirements:

• Open source derivative models, merged models, LoRAs, and products based on the above models.
• Share work details such as synthesis formulas, prompts, and workflows.
• Follow the fair-ai-public-license to ensure derivative works remain open source.

IV. Disclaimer

Generated models may produce unexpected or harmful outputs. Users must assume all risks and potential consequences of usage.