Checkpoint

A nice bland of fantasy and semi-realism 描述: 训练词语: ...

Hs UltraHD CG IllustriousIllustrious VersionRecommended SettingsParametersCFG: 5–7.5(may vary based on use case)Steps: 20–28Sampling Method: Euler aResolution: Total area aroun...

really unstable, but endearing... 描述: love you wrong time 训练词语: ...

really unstable, but endearing... 描述: never beating the shy boy allegations 训练词语: ...

really unstable, but endearing... 描述: Is the pizza 训练词语: ...

really unstable, but endearing... 描述: that one girl who's always doodling but never wants to show it to anyone 训练词语: ...

recommend to use with hires 描述: 训练词语: 名称: kh...

Base model dedicated to generating images of women of the time, rich and very powerful, with beautiful photographic outputs and well-detailed faces, has a high standard of beau...

hobohobo-Animagine+SPO+usuaji-stylehttps://civitai.com/models/769555/hobohoboanimaginehttps://civitai.com/models/510261/spo-sdxl4k-p10eplorawebuihttps://civitai.com/models/8376...

This model was aimed to generate anime images with anime coloring/screencap style. It was created by merging AnimagineXL 3.1 with my anime screencap LoRA and the DPO LoRA using...

该模型为 Artist All In One For NovelAI 1 LoRA 的大模型版本，使用该链接中的方法将 LoRA 模型融合进大模型中，模型使用方法请阅读 Artist All In One For NovelAI 1 LoRA 中的模型说明,T...

SLR PHOTO base model, designed to generate professional photographic images, with beautiful young models and women!I recommend using the Adetailer extension.Use this extension ...

Models here will likely be builds using ideas I have for broad spectrum improvements to Flux using low to no compute, usually using methods I call "back-of-van ablation," and m...

Compared to original WebUI (for SDXL inference at 1024px), you can expect the below speed-ups:If you use common GPU like 8GB vram, you can expect to get about 30~45% speed up i...

Crystal opal is a variety of opal that is highly prized for its transparent to translucent body and brilliant play-of-color, which appears to glow from within the stone. The te...

Realistic girls model, was implemented by a series of hybrid merges including data from loras sd 1.5 and loras flux dev 1 models, the hybrid result outperforms trained models a...

Compared to original WebUI (for SDXL inference at 1024px), you can expect the below speed-ups:If you use common GPU like 8GB vram, you can expect to get about 30~45% speed up i...

Models here will likely be builds using ideas I have for broad spectrum improvements to Flux using low to no compute, usually using methods I call "back-of-van ablation," and m...

Compared to original WebUI (for SDXL inference at 1024px), you can expect the below speed-ups:If you use common GPU like 8GB vram, you can expect to get about 30~45% speed up i...

fusion of gguf Q4.1 AND Q5.1 Flux.1 Lite 8B Alpha into one to be able to use it in forge. For me, Forge did not take the flux.1 lite 8B Alpha model.https://civitai.com/models/8...