Model
Trends
.ai
Model
Trends
.ai
the best open source ai models
Run Locally
beyondREALITY_v30.safetensors
11.08 GB · SafeTensor · fp8 · pruned
Downloads are served by Civitai. Some creators ask for a Civitai login first.
AD
Install / Run on Graydient
Requires membership
<edgeofreality3-flux>
About This Model
FLUX · Base Models
Created by
Nurburgring
on Civitai
Creator notes
SUPER Z image turbo rebuild 淡妆浓抹 V3.0 菁彩发布 20260210更新lora训练适配器 注意,是淡妆浓抹 3.0 模型的的lora训练专用,重点是diffusion,diffusion是新的训练才是新的,适配器是adapter,Rebuild版底模支持在此适配器上训练,可在aitoolkit中替换并微调自己专属lora。注意,不同于官方底模,是专用于本模型底模的训练适配器,感谢@Laushine制作,请改扩展名为ZIP后解压使用。 个人微调版 Z image turbo rebuild 淡妆浓抹 3.0发布 20260210补充FP8版 Z TURBO REBUILD , 更好的色彩光影表现,更细腻的纹理细节呈现。 比 Z image turbo 更多的艺术风格支持,比 Z image base 更快的生成速度。 CFG:1,10~15步, euler+simple 采样器调度器 善用SeedVarianceEnhancer等节点,增加画面的多样性表达 优化了一些艺术风格的支持与二次元动漫的表现。 Nunchaku量化版 by tonera(感谢): NVFP4量化以及采样器搭配 by miccodadada(感谢): 本人实测,AuraFlow3.5 - 5,首选euler a+FlowMatchEulerDiscreteScheduler,其次是ZImagePowerNodes自带的采样器,然后是DEIS_2M + BONG_TONGENT,追求泛用性的话是Sa_solver+FlowMatchEulerDiscrete,人像的话也可以试试ER_SDE+(sgm_uniform/FlowMatch)。 我自己制作了它的NVFP4混合量化,FEED FORWARD 123层采用NVFP4,其余注意力采用FP8 Per-Tensor量化(目前Comfyui暂时不支持BlockWise量化),质量下降在可接受范围内。在5070上,9步采样时间大概2.8-3.6秒。 欢迎各位试用 最后,团队继续招新,我一手搭建起Z image数据生产团队,目前因数据需求不断增加,Z image数据生产团队开启招新,坐标杭州某大厂,需要全职到岗。有意者加WX:Aptshtiso SUPER Z 淡妆浓抹 V2.0 菁彩发布 解决了一些已知问题 优化了纹理与细节表现,继续优化模型的高频细节表现。 可能拥有目前所有Z模型中,最佳的直出纹理细节表现能力,同时对胶片等摄影风格支持更佳。 同时尝试优化一些艺术风格的支持与二次元动漫的表现。 常规升级,欢迎试用,请多多返图~ 感觉比淡妆浓抹1.0版进步还是较为明显的(放大看细节表现) 为了获得多样性,请尝试SeedVarianceEnhancer节点 CFG:1.2 采样步数:10~15步 最后,团队招新,我一手搭建起Z image数据生产团队,目前因数据需求不断增加,Z image数据生产团队开启招新,坐标杭州某大厂,需要全职到岗。有意者加WX:Aptshtiso 欢迎加入我们Z image数据生产团队! ———————————————————————————————————————— SUPER Z Effortless Makeup - V2.0 - Vibrant Color Release Resolved some known issues The texture and detail presentation have been optimized, and the high-frequency detail presentation of the model is also being further improved. It may possess the best direct rendering texture detail performance capability among all current Z models, and also offers better support for photographic styles such as film. At the same time, efforts are made to enhance the support for various artistic styles and to improve the presentation of 2D anime. Regular upgrade. Welcome to try it out. Please share your feedback! The effect seems to have improved significantly compared to the previous version (viewing it in detail reveals this). To achieve diversity, please try the SeedVarianceEnhancer node. CFG: 1.2 Sampling steps: 10 to 15 steps Finally, the team is recruiting new members. I have established the Z image data production team by myself. Currently, due to the increasing demand for data, the Z image data production team has launched a recruitment drive. The location is a major company in Hangzhou. The position requires full-time employment. Those interested can add WeChat: Aptshtiso Welcome to join our Z Image data production team! ———————————————————————————————————————— SUPER Z 浓墨重彩 V1.0 菁彩发布 祝大家2026年元旦快乐~ 模型如其名,浓墨重彩 注意,浓墨重彩 版适合网红风格人像图片的生成,如果不喜欢这个风格,请使用浓妆淡抹1.0版。浓妆淡抹1.0版仅对纹理细节与质感做了优化。 这是一个网红偏好方向的微调模型,目前的浓墨重彩版本相比浓妆淡抹1.0版,继续优化纹理细节表现,纹理清晰,细节丰富,光影效果更加浓烈,搭配风格提示词效果较好,色彩表达大胆,偏胶片又不相同,略带磨皮效果,适合时尚写真类图片或社交网络风格图片。 同时尝试提升艺术化风格的表达能力,只是测试下来比较失败,艺术风格的支持还是不好。后续或许会专门出艺术风格支持良好的模型。 所以这是一个偏重口的高精度人像模型,请悉知 为什么搞这个模型?因为我发现各个讨论群中大量发布网红风格图,有些图很好看,但是细节表现并不好。这个模型如果你使用得好,你可以出很精彩的网图风格图片,如果控制不当,你会很失望。非模型之罪,风格取向不同。如果你觉得不好,尽管骂我~ 后续或许模型会分为两个分支: 浓妆淡抹,以提升纹理细节表现为主 浓墨重彩,以提升色彩光影表现为主 有用户反馈模型会有AI感,全看着自己的控制了哈哈哈 可以尝试在提示词前加一些风格描述,比如: 富士胶片风格 柯达胶片风格 徕卡镜头风格 蔡司镜头风格 电影风格 黑白默片风格 拍立得风格 ins风格 小红书风格 。。。。。。 你还可使尝试各种风格,或许会有不一样的惊喜 不要在意下载的模型名字还是浓妆淡抹,就是浓墨重彩,忘记改名了哈哈哈 SUPER Z - A Majestic and Colorful Grand V1.0 Release Wishing everyone a happy New Year's Day in 2026! The model is named as such, with rich and vivid details. Note: Version Majestic and Colorful Grand is suitable for generating images of popular style portraits. If you don't like this style, please use Version heavy and light makeup 1.0. Version heavy and light makeup 1.0 only optimized the texture details and texture quality. This is a micro-tuned model for the preferences of online influencers. Compared to version 1.0, the current version continues to optimize the representation of texture details. The textures are clear, the details are rich, the lighting effects are more intense, and the combination with style prompts works well. The color expression is bold, it leans towards film but is different from it, with a slight touch of smoothing effect. It is suitable for fashion photo-like images or social network-style images. At the same time, an attempt was made to enhance the expression ability of the artistic style, but the test results were not satisfactory. The support for artistic styles is still not good. In the future, a model with good support for artistic styles may be specially developed. So this is a high-precision portrait model with a pronounced facial feature. Please be informed. If you handle it well, you can create very impressive web-style pictures. But if you don't control it properly, you will be very disappointed. It's not the fault of the model; it's due to different style preferences. You can try adding some style descriptions before the prompt words, for example: Fuji Film style Kodak film style Leica lens style Zeiss lens style Film style Black-and-white silent film style Positivie style The Ins style Little Red Book style 。。。。。。 You can also try various styles, and perhaps you will have some unexpected surprises. SUPER Z 1.0或许是现在最佳的人像摄影 Z image 模型 注意最下面的参数设置。 好久不见,甚是想念 BEYOND REALITY系列模型断更已经半年多,主流开源绘图模型也已经快速迭代到了Z时代。 Z IMAGE给大家带来了足够的惊喜,超快的速度和超真实的画面,社区中对edit的呼声更是持续不断。 我知道你们很着急,但是先别急,先来看看社区微调版的 Z image 哈哈哈 BEYOND REALITY Z IMAGE 1.0 淡妆浓抹总相宜,这是一个基于人像数据集微调lora合并而来的模型,延续BEYOND REALITY系列模型的特点,高清晰、高美学。Z IMAGE目前的问题之一就是高频信息的表现不佳,体现在画面上就是纹理的细节不好。因此训练了一个lora来提升模型的纹理质量。本想发布lora的,后来想想,一直都是发布checkpoint的,那还是延续这个习惯吧。 我一直偏爱胶片摄影的光影色彩,因为那是无数美学工程师用百年的调色经验沉淀下来的美学的体现,是用图片记录光影色彩、记录美的最佳的、受众最广泛的体现。美不是被定义的,而是被感受的,能被定义的或者量化的,一定是会沦为平庸。所以美无法量化无法评估,只能去理解和感悟。美也无法让所有人都共情,能遇到一个人理解你,已经是万幸之事。 本次更新主要优化Z的人像类皮肤纹理和环境纹理细节,并优化模型生成人像类图像的美学。Z 本身生成的图片真实度非常高,某些情况下会让你完全分不清是AIGC图还是真实照片。但是,在美学方面,Z 还有优化提升的空间。 本次更新的模型整体风格略偏胶片人像摄影,可以生成很美的人像类图片,也可以生成很优美的环境细节,同时保留了 Z 的部分真实感。在提示词引导下可以生成非常真实的图片。 注意,因本次训练以合成的高质量人像摄影风格图
3,736 downloads · 424 likes on Civitai
SHA256
27EF57F6498DCF9FBA30B46F0E2ABFDE2D45CCA20D410D3140C8E1F68CAD7A1F
ModelTrends.ai Model ID
#16951
Flag content
Example renders
Prompts and settings shared by the people who made these renders on Civitai. Pick a render to see how it was made.
Settings
Sampler
Euler
Steps
20
Guidance
1
Seed
2579717183
Size
1080x1350
Prompt
A young woman, likely in her late teens or early twenties, of East Asian ethnicity, is seated in a vibrant pink and coral-colored, ornate, chair-like structure. She is positioned slightly off-center to the left side of the image. The woman is oriented directly facing the viewer, with a pleasant expression. She has long, dark brown hair, and a light complexion with defined, delicate features. Her posture is relaxed and casual. She is wearing a light pinkish-beige, patterned blazer-style jacket over a white top and a black pleated mini skirt. She is wearing a light pink headband with bunny ears, a small, plush, pink purse-shaped accessory, and a wristwatch. Her body type is slender and her legs are visible. The environment is a playful, stylized, and bright interior space, almost resembling a whimsical themed cafe or shop. The decor is primarily pink, coral, and cream, featuring intricate patterned details on the chair. The lighting is well-balanced and soft, highlighting her features without harsh shadows. The overall style is cute, cheerful, and a bit whimsical. The composition is straightforward, focusing on the subject and her interaction with the decor. Important details include the cartoon-like purse, and the overall pastel-colored themed ambiance. <lora:korean_neurocanvas_v1:1>
Settings
Sampler
Euler
Steps
20
Guidance
1
Seed
674555672
Size
896x1152
Prompt
(photorealistic,4k,8k,masterpiece,best quality:1.4),(Japanese beautiful,woman,early twenties,fashion model,detailed face and eyes,very beautiful face,blue eyes,double-edged eyelid:1.4),(oval face,small face,slim waist:1.4),(sharp eyeblows,delicate chin,delicate ears,Roman nose:1.4),(natural makeup,pink lips,thin lips,smirk:1.2),(downturned eyes,big eyes,fit,big breast:1.4),(Tall,long legs,Healthy and fair skin:1.4),(hair over one eye,short bob cut,pink hair:1.4)
AI Models for Image, Video & Text Generation
edgeofreality3 flux is a base model for the Flux family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Flux models, check its heat score to see how it is trending, and open it on PirateDiffusion or BitVector to try it.
Browse by Model Family
Browse all models