--- license: other license_name: faipl-1.0-sd license_link: https://freedevproject.org/faipl-1.0-sd/ language: - en tags: - text-to-image - stable-diffusion - safetensors - stable-diffusion-xl ---
Raehoshi illust XL 3
## Overview Introducing **Raehoshi illust XL 3** , an enhanced iteration built upon the IllustriousXL model. It aims to elevate the visual style by addressing some of the limitations in the original, such as oversaturation and artifact noise. While these issues are not entirely eliminated, noticeable improvements have been made. The goal is to deliver a more polished, balanced output while staying true to the strengths of the base model. ## Model Details - **Developed by**: [Raelina](https://civitai.com/user/Raelina) - **Model type**: Diffusion-based text-to-image generative model - **Model prompt style**: Booru-tags - **License**: [Fair AI Public License 1.0-SD](https://freedevproject.org/faipl-1.0-sd/) - **Finetuned from**: [Illustrious XL v0.1](https://huggingface.co/OnomaAIResearch/Illustrious-xl-early-release-v0) ## Recommended settings - **Positive prompts:** ``` masterpiece, best quality, very aesthetic, ``` **Special tags:** "dramatic lighting" - **Negative prompts:** ``` lowres, bad quality, worst quality, bad anatomy, sketch, jpeg artifacts, signature, watermark, ``` - **CFG:** 5-6 - **Sampling steps:** 28 - **Sampler:** Euler a - **Supported Resolution:** ``` 1024 x 1024, 1152 x 896, 896 x 1152, 1216 x 832, 832 x 1216, 1344 x 768, 768 x 1344, 1536 x 640, 640 x 1536 ``` ## Hires.fix Setting - **Upscaler:** [4x_NMKD-YandereNeoXL](https://nmkd.de/?esrgan) - **Hires step:** 10-15 - **Denoising:** 0.1-0.3 ## Training config The model was developed using a two-stage fine-tuning process. In Stage 1, new series and characters were introduced into the model. Stage 2 focused on fixing issues and enhancing the overall style for improved output. Stage 1 - Dataset : 34k - Hardware : 2x H100 80gb - Batch size : 32 - Gradient accumulation steps : 2 - Learning rate : 5e-6 - Text encoder : 2.5e-6 - Epoch : 10 Stage 2 - Dataset : 2.3k - Hardware : 1x A100 80gb - Batch size : 48 - Gradient accumulation steps : 1 - Learning rate : 3e-6 - Text encoder : disable - Epoch : 10 ## License [Fair AI Public License 1.0-SD](https://freedevproject.org/faipl-1.0-sd/)