AI Tools

Feel free to try this latest SD3 Medium basic workflow!

提示词参考：american Shorthair，no humans,animal focus, looking at viewer, anime artwork, anime style, key visual, vibrant, studio anime, highly detailed，StdGBRedmAF,Studio Ghibli,对于宠物体态 样貌有不错的还原，但对于宠物自身的颜色无法做到一比一还原。

支持表情包人物一键抠图

可图大模型是由快手可图团队开发的基于潜在扩散的大规模文本到图像生成模型。Kolors 在数十亿图文对下进行训练，在视觉质量、复杂语义理解、文字生成（中英文字符）等方面，相比于开源/闭源模型，都展示出了巨大的优势。同时，Kolors 支持中英双语，在中文特色内容理解方面更具竞争力。

上传一张图片即可自动移除背景，输出处理之后的图片以及遮罩

Typer the superhero character that you want to generate!（The overall style was inspired by user Shunkaha）

上傳圖片 (Load image)簡單描述 (Prompt)描述控制強度 (CFG)貼近原稿程度 (strength of close to original image)張數 (batch size)垂直元素 (Vertical Decoration: Highlighting columns, vertical decorative walls, and panels.)強度 (strength)水平元素 (Horizontal Decoration: Emphasizing beams, balconies, and cantilevers.)強度 (strength)框架元素 (Frames: Combining the above two to form a frame.)強度 (strength)挖空量體 (Void volumes: Creating voids or cavities within masses.)強度 (strength)堆疊量體 (Stacked volumes: Accumulating masses to create a layered effect.)強度 (strength)

※ V1.2 : Added radio button promptsLoRA × ２ と アップスケール が可能な仕様です* By default, two images are created at a time.　You can change the number of images with "batch_size".* The default settings use the prompts and settings of Creator: Kei.

Upscaler for all images Note : its a little slow but the result is worth it.Tip : use low resolution images upto 1000px to get fast result

Stylize your AI-created images or your photos with this filter.

HHM Figure ground siteplan

Please upload a clear, front-facing photo of the face you want to transform and enter the prompt so the AI Tool can generate a new image with your original face.

Sometimes you generate a great image using SD3, but upon closer inspection, the face turns out to be a nightmare! In that case, use my WF for a quick fix ;)Here is a free SD3 checkpoint: Aderek SD3 - v1 | Stable Diffusion Model - Checkpoint | Tensor.Art

Change Output Flat Strength: from 0.5 to 0.7 to produce results that are similar to the original image. Below 0.5 the results will be similar or unchanged from the original image. Above 0.75 the generated image will be very different from the original image. Default is 0.6 - 0.65Change Flat Ratio Strength: to apply flat style more or lessExtra Detailer: to add more details to the resulting image

Ink art style image generation at very low cost (around 0.6). Radio button based user prompt input is tried.Default prompts are given for ink art settings and Negative prompts. Just fill the character and scene related descriptions by selection or your own prompt. Enjoy~

Building On fire

A cool spot to generate your work with Loras

上傳圖片 (Load image)簡單描述 (Prompt)描述控制強度 (CFG)貼近原稿程度 (strength of close to original image)張數 (batch size)垂直元素 (Vertical Decoration: Highlighting columns, vertical decorative walls, and panels.)強度 (strength)水平元素 (Horizontal Decoration: Emphasizing beams, balconies, and cantilevers.)強度 (strength)框架元素 (Frames: Combining the above two to form a frame.)強度 (strength)挖空量體 (Void volumes: Creating voids or cavities within masses.)強度 (strength)堆疊量體 (Stacked volumes: Accumulating masses to create a layered effect.)強度 (strength)

簡單描述 (Prompt)描述控制強度 (CFG)張數 (batch size)垂直元素 (Vertical Decoration: Highlighting columns, vertical decorative walls, and panels.)強度 (strength)水平元素 (Horizontal Decoration: Emphasizing beams, balconies, and cantilevers.)強度 (strength)框架元素 (Frames: Combining the above two to form a frame.)強度 (strength)挖空量體 (Void volumes: Creating voids or cavities within masses.)強度 (strength)堆疊量體 (Stacked volumes: Accumulating masses to create a layered effect.)強度 (strength)

Rainning

Anime style character generator (e.g., Azur Lane)

value 0.2-0.3 : try again to fix fingervalue 0.4-0.5 : this way, go aheadvalue 0.6 : balance image to imagevalue 0.7-0.8 : give me a image like this stylevalue 1.0 : text to image, only use image aspect

The tool will produce 2 results, with the best effect depending on each different case:- Image 1: the product image will have a slight blur at the edge- Image 2: the product image will not have a blur at the edge

EK-Realism Art [SD3] creates images with realism and details as well as beauty.or easy usage, radio button prompt UI is used~ 🤗😉VAE: nonesampler: euler, DPM++ 2Mscheduler: normal, SGM UniformLora strength: 0.5~1.0Ksampler steps: 25+cfg: around 7denoise: around 0.5upscaler: 4x-UltraSharp,R-ESRGAN 4x+Currently, because of SD3 scaler bug, I don't use scaler until it is fixed.

ONE IMAGE PER ⚡​0.8UNA IMAGEN POR ⚡​0.8

上傳圖片 (Load image)簡單描述 (Prompt)描述控制強度 (CFG)貼近原稿程度 (strength of close to original image)張數 (batch size)垂直元素 (Vertical Decoration: Highlighting columns, vertical decorative walls, and panels.)強度 (strength)水平元素 (Horizontal Decoration: Emphasizing beams, balconies, and cantilevers.)強度 (strength)框架元素 (Frames: Combining the above two to form a frame.)強度 (strength)挖空量體 (Void volumes: Creating voids or cavities within masses.)強度 (strength)堆疊量體 (Stacked volumes: Accumulating masses to create a layered effect.)強度 (strength)

open workflow to change more parameters

open workflow to change more parametersIMPORTANT1 = Non Face | 2 = Portrait with Face

A workflow that is creating vector-like art by 2 steps. The first step creates the "base" picture, then using this picture creates a vectorized version.

This version improves on the original with a wider variety of styles.Simply type in the name of the character you want.If you have any suggestions feel free to leave your comments.

簡單描述 (Prompt)描述控制強度 (CFG)張數 (batch size)垂直元素 (Vertical Decoration: Highlighting columns, vertical decorative walls, and panels.)強度 (strength)水平元素 (Horizontal Decoration: Emphasizing beams, balconies, and cantilevers.)強度 (strength)框架元素 (Frames: Combining the above two to form a frame.)強度 (strength)挖空量體 (Void volumes: Creating voids or cavities within masses.)強度 (strength)堆疊量體 (Stacked volumes: Accumulating masses to create a layered effect.)強度 (strength)