<h1 id="chocho-butterfly-mode-v2-325jmvyig">Chocho Butterfly Mode v2</h1>Follow me on Bluesky @<a target="_blank" rel="ugc" href="http://burnthevillage.bsky.social">burnthevillage.bsky.social</a>: <a target="_blank" rel="ugc" href="https://bsky.app/profile/burnthevillage.bsky.social">https://bsky.app/profile/burnthevillage.bsky.social</a>2/14/2025 🐴 Pony Update: Happy Valentines day! Added pony model that is still a work in progress BUT still Much better than previous models. I predict the Final version will be pretty different than this model because plenty things to improve when it comes to training settings and cleaning the dataset but it can produce some decent photos with pony so I didn't want to delay its release any further. Feedback encouraged!update! - 06/06/2024: Drastically improved this lora with various techniques to make the character more accurate and easier to prompt for in general. I used different captioning when training so the tags have slightly changed but for the better! The only new clothing I added is for the yellow dress which is now able to be worn with sandals (<a target="_blank" rel="ugc" href="https://i.kym-cdn.com/entries/icons/original/000/025/376/DJrmQ8yV4AA9JjX.jpg">Hell fuckin' Yeah!!!!!!</a>).Key Words:<ul><li>Trigger words: chocho akimichi, woman</li><li>With wings: butterfly mode, purple wings,</li><li>Default outfit: chodef, forehead protector, white kimono, fishnet shirt, toeless footwear, toes, arm warmers, sleeveless, yellow sash, dark yellow ribbon,</li><li><a target="_blank" rel="ugc" href="https://naruto.fandom.com/wiki/Super_Ch%C5%8Dch%C5%8D_Kiss_Mode!?file=Ch%C5%8Dch%C5%8D_Kiss_Mode.png">Yellow dress</a> from "<a target="_blank" rel="ugc" href="https://naruto.fandom.com/wiki/Ch%C5%8Dch%C5%8D_Arc">Chōchō Arc</a>": chodress, yellow choker, yellow dress, orange obi, orange bow, (sandals), nervous expressions, sweat drops,</li><li>Framing the shot: hand on hip, fully body, pointing, upper body, close-up</li></ul>Suggested settings:<ul><li>The eyes on this one were tricky so I got lazy and used <a target="_blank" rel="ugc" href="https://civarchive.com/models/178518/eyeful-or-robust-eye-detection-for-adetailer">adetailer for the eyes only</a> where I used <a target="_blank" rel="ugc" href="https://civarchive.com/models/120723">DetailedEyes_XL</a> inside the adetailer positive prompt (Exact adetailer metadata you can get from the "copy generation data" button). On the flip side, I didn't use tools like Controlnet, face restore or Inpainting. Outputs are typically just what you would get from putting in text and using a high-res fix, which I suggest a <a target="_blank" rel="ugc" href="https://civarchive.com/models/147821/realesrganx4plus-anime-6b">4x upscaler</a> and a little adetailer.</li><li>I had CLIP skip 2 on in all pics, I trained the LORA with Clip Skip 2.</li><li>I started using <a target="_blank" rel="ugc" href="https://github.com/lllyasviel/stable-diffusion-webui-forge">Forge Webui</a> A fork of <a target="_blank" rel="ugc" href="https://github.com/AUTOMATIC1111/stable-diffusion-webui">AUTOMATIC1111</a>. It comes with new samplers and I used Euler a Turbo, but regular Euler a should be fine I just prefer the <a target="_blank" rel="ugc" href="https://www.youtube.com/watch?v=l71ZCs7tUu8">turbo</a>.</li></ul>I only did testing with <a target="_blank" rel="ugc" href="https://civarchive.com/models/84586/aam-anylora-anime-mix-anime-screencap-style-model">AAM - AnyLoRA Anime Mix</a>, but I trained with <a target="_blank" rel="ugc" href="https://civarchive.com/models/260267/animagine-xl-v31">Animagine XL V3.1</a> as a base so I'm hoping it should work with most models like my last few loras. Please feel free to leave any feedback!love yah,Florida-Man<h1 id="chocho-butterfly-mode-v1-0n26x5fcp">Chocho Butterfly Mode v1</h1><h2 id="story-v3ykiw83x">Story</h2>Hello, Thanks for stopping by! I started training this model last year since there were no Chocho models at the time. I used it to learn how to train LORA's on SD1.5 but was never really satisfied with it, so I never uploaded it. I decided to continue working and make an SDXL version of the model but I do plan to release a SD1.5 version as well. Even to this day (At least from my limited search), there's still no models with Chocho in her butterfly mode, so I think this is the first one out there ;). So I'm excited to finally bring this to the world and I think I got some super cool results. Unfortunately, I already see where I can improve the model, so there will be another version in the future. But I'm tired now :'(.Key Words:<ul><li>With wings: butterfly_wings,</li><li>Clothing Prompt: double buns, white sash, orange ribbon, black arm warmers, fishnet, (Note: This should should only be necessary if you are using something other than, "<a target="_blank" rel="ugc" href="https://civarchive.com/models/269232/aam-xl-anime-mix?modelVersionId=303526">AAM XL (Anime Mix)</a>" by Lykon. Try placing in positive / negative prompt to add or change her drip).</li><li>Full body: hand on hip, fully body, pointing,</li><li>Yellow dress from "Chōchō Arc": yellow choker, yellow dress, nervous expressions, sweat drops,</li></ul>Suggested settings:<ul><li>First things first, I wanted the pictures to come off as natural as possible. so I didn't use certain tools like ADetailer, Controlnet, face restore or Inpainting. Outputs are typically just what you would get from putting in text and using a high-res fix, which I suggest a 4x upscaler.</li><li>I had CLIP skip 2 on in some pics, I trained the LORA with Clip Skip 2.</li></ul>I tested with a few different XL models including a lightning model which I included some decent results. I only did a few hours of testing however, so please feel free to leave any feedback! Have Fun,Florida-ManOfficial HF repository: (Soon)<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/ba0da5f1-bf40-46b5-88b3-1291faa5c729/width=525/ba0da5f1-bf40-46b5-88b3-1291faa5c729.jpeg" /><a target="_blank" rel="ugc" href="https://ko-fi.com/904floridaman"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/57449acc-1150-48ba-ad0f-e6d3cb84ebd5/width=525/57449acc-1150-48ba-ad0f-e6d3cb84ebd5.jpeg" /></a>Your contribution is greatly appreciated and helps me to create more content. If you like my model, please consider donating: <a target="_blank" rel="ugc" href="https://ko-fi.com/904floridaman">https://ko-fi.com/904floridaman</a>I'm open to commissions.

Chocho Akimichi - Butterfly Mode (秋道チョウチョウ, Akimichi Chōchō)

<ul><li>Improved overall quality of model</li><li>More accurate skin color</li><li>More consistent character and outfits</li><li>Easier to frame the picture</li><li>Better captioning for easier tagging</li></ul>

Chocho_Akimichi_sdxl_v2.safetensors

Chocho Akimichi - Butterfly Mode (秋道チョウチョウ)

SDXL v2.0

(beta) Pony RC4

SDXL v1.0

Key Words:<ul><li>Trigger words: chocho akimichi, woman</li><li>With wings: butterfly mode, purple wings,</li><li>Default outfit: chodef, forehead protector, white kimono, fishnet shirt, toeless footwear, toes, arm warmers, sleeveless, yellow sash, dark yellow ribbon,</li><li><a target="_blank" rel="ugc" href="https://naruto.fandom.com/wiki/Super_Ch%C5%8Dch%C5%8D_Kiss_Mode!?file=Ch%C5%8Dch%C5%8D_Kiss_Mode.png">Yellow dress</a> from "<a target="_blank" rel="ugc" href="https://naruto.fandom.com/wiki/Ch%C5%8Dch%C5%8D_Arc">Chōchō Arc</a>": chodress, yellow choker, yellow dress, orange obi, orange bow, (sandals), nervous expressions, sweat drops,</li><li>Framing the shot: hand on hip, fully body, pointing, upper body, close-up</li></ul>Suggested settings:<ul><li>The eyes on this one were tricky so I got lazy and used <a target="_blank" rel="ugc" href="https://civitai.com/models/178518/eyeful-or-robust-eye-detection-for-adetailer">adetailer for the eyes only</a> where I used <a target="_blank" rel="ugc" href="https://civitai.com/models/120723">DetailedEyes_XL</a> inside the adetailer positive prompt (Exact adetailer metadata you can get from the "copy generation data" button). On the flip side, I didn't use tools like Controlnet, face restore or Inpainting. Outputs are typically just what you would get from putting in text and using a high-res fix, which I suggest a <a target="_blank" rel="ugc" href="https://civitai.com/models/147821/realesrganx4plus-anime-6b">4x upscaler</a> and a little adetailer.</li><li>I had CLIP skip 2 on in all pics, I trained the LORA with Clip Skip 2.</li><li>I started using <a target="_blank" rel="ugc" href="https://github.com/lllyasviel/stable-diffusion-webui-forge">Forge Webui</a> A fork of <a target="_blank" rel="ugc" href="https://github.com/AUTOMATIC1111/stable-diffusion-webui">AUTOMATIC1111</a>. It comes with new samplers and I used Euler a Turbo, but regular Euler a should be fine I just prefer the <a target="_blank" rel="ugc" href="https://www.youtube.com/watch?v=l71ZCs7tUu8">turbo</a>.</li></ul>I only did testing with <a target="_blank" rel="ugc" href="https://civitai.com/models/84586/aam-anylora-anime-mix-anime-screencap-style-model">AAM - AnyLoRA Anime Mix</a>, but I trained with <a target="_blank" rel="ugc" href="https://civitai.com/models/260267/animagine-xl-v31">Animagine XL V3.1</a> as a base so I'm hoping it should work with most models like my last few loras. Please feel free to leave any feedback!

Files with SHA256: 45291ec129da83109f6a44a087dd816a7c66914d47bb8ce0fc1d4b27a270ae39