Checkpoint

V1.0版本(ytpony3860XL-014):分辨率:832×1216 768×1152采样方式:Euler a迭代步数:20到30之间高清修复步数:20到30之间CFG:6左右重绘幅度:0.20、0.45放大算法:R-ESRGAN 4x+ Anime6B可参考示例...

If you liked this...Try UltronMerge!UltronMerge - UltronMerge | Stable Diffusion Checkpoint | CivitaiI consider it to be an improvement over this model and with all the same pe...

ImportantModels uploaded here are for only eveluation purpose, but it is free to generate the picture.It might be uploaded with official model name with the optimization.Also, ...

This is a model that I just liked while making the merge model.Please try using it if you think I might like it too. 描述: 训练词语: ...

Get Leek'd!Leek is a merge intended to bring style and characters back into the user's hands, without relying on underbaked style LoRAs that end up working on a single model.Re...

Check my models on Mage and get unlimited generations: ConsistentFactor /Consistent Factor Euclid /Consistent Factor Borealis /Consistent Factor Lumina /Consistent Factor Allur...

My first Pony merge !It's the result of mostly AIOS V101Foal + RAS - RealAnimeScreencap on some layers with some style loras sprinkled on topI do like the style I ended up with...

The 1.5 version of Colorful Fantasy Checkpoint is a combination of Colorful Eurasian Girl 彩璃混血女孩, Colorful Anime Kawa 彩璃二次元, Twin-Ponytail Girl and some other models...

Just my junky models I use to fix up my MJ/Niji imagesI merge a bit...maybe a bit too much ? and sometimes I mix models that I end up liking, so I thought I would share,Will th...

Colorful Realistic XL, originally driven by personal interest, has been fine-tuned using a total of 12,379 high-quality images, tagged with GPT4V, WD14, and COGVLM. This versio...

This model is good at creating visual novel style illustrations, with detailed character and high-quality background.Prompts for models derived from Anything work well on this ...

Elite is a high standard model, it is not just any model and do not judge it because the name is commercial, it will generate photographic images and general purpose, quality a...

This model is for creating portrait shots. It's a mix of realisticStockPhoto_v30, leosamsFilmgirlUltra and subredditV7 描述: 训练词语: ...

0002v3+0002Pony_v2ALTERNATIVESTYLE23.5 = 0,0,0,0,0,0,0,1,1,1,0,0,1,1,1,1,1,0,0,0 = Satan 描述: v2 merges 2 loras, sakimori and quasarcake with 0.25 weight eac...

Young cute girls. The model integrates some loras that modify the body.Recommended for a better result: ADETAILER mediapipe face mesh. VAE is included.Setting : Sampling Euler-...

Are you looking for a model that's both 'Realistic'? Try Lemix Realistic Asians!全能的亚洲真人、真物、真景写实模型，喜欢可以试一下！★ Please upload at least one photo made usin...

Are you looking for a model that's both 'Realistic'? Try Lemix Realistic Asians!全能的亚洲真人、真物、真景写实模型，喜欢可以试一下！★ Please upload at least one photo made usin...

采样方法DPM++ 2M KarrasEuler a步数30负面词(worst quality, low quality),deformed,distorted,disfigured,doll,poorly drawn,bad anatomy,wrong anatomy, 描述: ...

UrangDiffusion v2.5 (oo-raw-ng Diffusion) brings a whole-new training method compared to the v2.0. The model provide more flexibility and brings some updated dataset.The name “...

Matrix LightningMatrix Lightning is a merge of Mangled Merge Matrix into the SDXL Lightning 4 step model released upon request. I recommend using the following settings based o...