在tashan real pony的基础上进行调整训练，使出图人物更加偏向熟女。Based on the "tashan real pony" model, we have conducted further fine-tuning to generate images of characters that appear more mature.现在仅是测试版，仅完成了身材调整，真实感及亚洲脸还未完成This is currently a test version, and only body shape adjustments have been completed. Enhancements for realism and Asian facial features are still in progress.大家在用的时候有什么问题可以在评论区反馈，我会在能力范围内解决。 If you encounter any issues while using it, please feel free to leave feedback in the comments section, and I will address them within my capabilities.食用说明：1.正向提示词：score_9,score_8_up,score_7_up,hd,8k,masterpiece,high quality,realistic,a female,不建议使用1girl,使用a female效果更好2.负面提示词： score_6,score_5,score_4,(worst quality:2),(low quality:2),(normal quality:2),bad anatomy,bad hands,missing fingers,fewer digits,source_furry,source_pony,source_cartoon,3d,blurry,watermark,logo,3.采样步数在20步时画面已趋近完美，30步之后无收益，建议25-30。4.vae已内置vae,不用额外选择。5.高分辨率修复及ADetailer半身照及大头照没有开的必要，全身照看情况开，远景照建议全开画面如有些许崩坏，重绘幅度给到0.6，可以在一定程度解决问题，其它无崩坏的建议不要超过0.46.采样器选择建议使用euler a。7.初始出图尺寸建议在7681024以上。Usage Instructions：Positive Prompt Words: score_9, score_8_up, score_7_up, hd, 8k, masterpiece, high quality, realistic, a femaleIt is not recommended to use "1girl"; using "a female" yields better results.Negative Prompt Words: score_6, score_5, score_4, (worst quality:2), (low quality:2), (normal quality:2), bad anatomy, bad hands, missing fingers, fewer digits, source_furry, source_pony, source_cartoon, 3d, blurry, watermark, logoSampling StepsThe image tends to be near perfect at around 20 steps; beyond 30 steps, there are no gains. It is suggested to use 25-30 steps.VAEVAE is built-in and does not need to be selected additionally.High-Resolution Fix and ADetailerThere's no need to enable these for half-body shots or close-ups. For full-body shots, it depends on the situation, while for distant shots, it's recommended to turn them on fully.If the image has slight issues, setting the redraw magnitude to 0.6 can resolve the problem to some extent. For images without such issues, it is advised not to exceed 0.4.Sampler SelectionIt is recommended to use "euler a".The initial image size should preferably be 7681024 or larger.免责声明一、使用责任1.您基于本模型服务生成的文本内容由您自行维护并独立判断后使用。您需在遵守适用法律法规和本服务条款的条件下使用。2.基于本服务生成的内容产生的任何知识产权问题由您自行处理，您需对生成内容负责。我们不对由此造成的任何损失承担责任。二、知识产权3.除法律法规另有规定外，未经我们书面授权同意，不得将本模型服务集成在任何第三方软件或服务中。4.亦不得以任何直接或间接、明示或暗示的方式向其他任何第三方披露生成内容与本服务的关联性。三、商业用途限制5.未经我们书面授权同意，非个人用户（包括但不限于企业、盈利性组织）不得将本模型用于商业用途（包括但不限于文字生成图像、图像生成图像、换脸等）。6.如您违反本条约给我们造成的损失，我们有权向您追偿。DisclaimerI. Usage ResponsibilityThe text content you generate based on this model service is maintained by you and used at your own discretion following independent judgment. You must use the service in compliance with applicable laws, regulations, and the terms of this service.Any intellectual property issues arising from the content generated through this service are to be handled by you, and you are responsible for the generated content. We do not assume any responsibility for losses incurred as a result.II. Intellectual Property RightsUnless otherwise provided by laws or regulations, without our written authorization and consent, you may not integrate this model service into any third-party software or services.You also must not disclose the relationship between the generated content and this service to any other third party in any direct or indirect, explicit or implicit manner.III. Restrictions on Commercial UseWithout our written authorization and consent, non-individual users (including but not limited to businesses, profit-making organizations) shall not use this model for commercial purposes (including but not limited to text-to-image generation, image-to-image generation, face swapping, etc.).If you violate this agreement and cause losses to us, we have the right to seek compensation from you.
