Stable Diffusion 2.1 anime model. Based on a mix of Plat Diffusion, PRMJ, CModel. Then trained on a dataset of quality anime art. The Waifu Diffusion 1.4 kl-f8-anime2 VAE is recommended (already baked into the model).<ul><li>To improve quality add to start of prompt: masterpiece, best quality, high quality, absurdres.</li><li>Example negative prompt: (worst quality:1.2), low quality, medium quality, deleted, lowres, comic, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, jpeg artifacts, signature, watermark, username, blurry</li></ul>

Mishi Anime

Mishi_Anime_V1.yaml

Mishi_Anime_V1.safetensors

For more information, <a target="_blank" rel="ugc" href="https://huggingface.co/hearmeneigh/sd21-e621-rising-v1">see here</a>.<ul><li>Guaranteed NSFW or your money back</li><li>Fine-tuned from <a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-2-1-base">Stable Diffusion v2-1-base</a></li><li>19 epochs of 450,000 images each, collected from <a target="_blank" rel="ugc" href="https://e621.net/">E621</a> and curated based on scores, favorite counts, and tag filtering.</li><li>Trained with <a target="_blank" rel="ugc" href="https://huggingface.co/hearmeneigh/sd21-e621-rising-v1/blob/main/meta/tag-counts.json">5,356 tags</a></li><li><code>512x512px</code></li><li>Compatible with 🤗 <code>diffusers</code></li><li>Compatible with <code>stable-diffusion-webui</code></li><li>Likely compatible with anything that accepts <code>.ckpt</code> and <code>.yaml</code> files</li></ul>

E621 Rising: A Stable Diffusion 2.1 Model

sd21-e621-rising-v1-epoch19.yaml

sd21-e621-rising-v1-epoch19.ckpt

Epoch 19

This model is dreambooth trained based on SD2.1-512. Dataset consist of multiple glitched imagesPreviously released here: <a target="_blank" rel="ugc" href="https://huggingface.co/sd-dreambooth-library/glitched">https://huggingface.co/sd-dreambooth-library/glitched</a>You can support me and request new lora's/checkpoints at: <a target="_blank" rel="ugc" href="https://boosty.to/smgbuddies/donate">https://boosty.to/smgbuddies/donate</a>

Glitched

glitched-v1.yaml

glitched-v1.safetensors

v1.0

This is dreambooth from SD2.1-512 base model. Dataset consist of multiple weirdcore images, that I selected myself. Previously posted at: <a target="_blank" rel="ugc" href="https://huggingface.co/sd-dreambooth-library/weirdcore">https://huggingface.co/sd-dreambooth-library/weirdcore</a> You can support me and request lora/checkpoints at: <a target="_blank" rel="ugc" href="https://boosty.to/smgbuddies/donate">https://boosty.to/smgbuddies/donate</a>

Weirdcore

weirdcore.safetensors

weirdcore.yaml

Retro3d dreambooth based on SD2.1-512. Dataset contains some old school games screenshots like: blood2, silent hill, mario64, final fantasy and other games from 90s and early 00sAs far as I can see, it's not very good in holding the prompt - sometimes it tends to ignore it in favour of staying in style. But! I think it's very creative and good in terms of making new stuff. I like to think of this model as a creative designer who isn't very good at following directions but creates very interesting stuff. previously served here: <a target="_blank" rel="ugc" href="https://huggingface.co/sd-dreambooth-library/retro3d">https://huggingface.co/sd-dreambooth-library/retro3d</a>You can support me and request new lora's/checkpoints at: <a target="_blank" rel="ugc" href="https://boosty.to/smgbuddies/donate">https://boosty.to/smgbuddies/donate</a>

Retro3D

Reuploaded model. Previously was uploaded as 1.5, but it was wrong. It was always based on sd2.1

Files with SHA256: b2379501101f1dd2c7eb1e50d24dedace0378048ab3aef4c78a434944785b044