LoRA trained on the Nami (Post Time Skip) Outfit from from One Piece
nami cosplay, 1girl, bikini, jeans
anime: MeinaMix - Meina V11, YesMix - v3.5, DivineEleganceMix v7
2,5D: irRealitium, Aniverse, Dark Sushi 2.5D, GhostMix - v2.0, Animerge - V2.3
better hands LoRA: EnvyBetterHands LoCon - beta2
negative embeddings: Async's Negative - v11-light, veryBadImageNegative, Bad artist
Recommended settings:
Make sure you are aware on the usage instructions of LORA
VAE: vae-ft-mse-840000-ema-pruned (for realistic models & 2,5D), kl-f8-anime2, blessed2.vae (for anime models)
Sampler: DPM++ SDE Karras or DPM++ 2M Karras(euler a for img2img upscaling) from time to time im also using DPM++ 2M alt Karras
Steps: 20 to 35
Clipskip: 2
Upscaler : GitHub - pkuliyi2015/multidiffusion-upscaler-for-automatic1111 4x-UltraSharp(for 2,5D & realistic), 4xAnimesharp, RealESRGAN_x4plus_anime_6B.pth
CFG Scale : 5 to 9 (mostly 7)
LORA weight for txt2img: 0.5-1 (for the examples i mostly used 0.8 and 1)
Denoise strength upscaling: 0.3 to 0.55
If you want to autocomplete the tags & trigger im using, I recommend using Booru style tag autocompletion for AUTOMATIC1111's Stable Diffusion web UI
For full body shots i recommend using: ComfyUI Impact Pack
or the a1111 alternative GitHub - Bing-su/adetailer
Description
nami cosplay, 1girl, bikini, jeans
FAQ
Comments (3)
What's your method of training for separating outfits like this from the character? Are there just enough images out there of characters other than Nami wearing this outfit that you can train on a variety of characters? Or do you mostly start with images of Nami wearing the outfit then train on AI-generated images in some way?
Some parts feature images of other characters wearing Nami's outfit, while other parts contain fan art where the faces look less like Nami. If this option isn't available, you can also cover the faces with black bars in some images
There is also a great guide for that: An Easy Way to Make A Cosplay Lora 简易Cosplay Lora制作法 | Civitai
@AhriMain Thanks!
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.