Description:
Made with a local model called EasyFluff:
SD install guide:
https://rentry.org/voldy
Model:
https://huggingface.co/zatochu/EasyFluff/tree/main
Grab these from hugging face and put them in the automatic\models\Stable-diffusion folder:
EasyFluffV11.2.safetensors
EasyFluffV11.2.yaml
And download this vae: https://civitai.com/models/23906/kl-f8-anime2-vae
And put it in the automatic\models\Vae folder.
After you open SD in your browser go to the extensions tab and download this extension with this link: https://github.com/Seshelle/CFG_Rescale_webui then restart it.
- Natural language has a strong impact on this model as of V7. It won't understand everything but it can get you places. And then when you can't squeeze any more out of it, you can use e621 tags to help refine your prompt. (or you can just use tags, either way will work.)
- Start your prompt with a short natural language prompt describing what you want, then pad it with e621 tags to refine specific concepts. As this is a v-prediction model, prompt interpretation can be a bit more literal.
- Many flavor words and artists from SD 1.5 work again.
- PolyFur is trained on MiniGPT-4 captions, so try being really flowery with your prompts and use sentences even. (check the generation info on the preview images to see how)
- If an established character isn't coming out accurately, try increasing the strength of their token and adding a few implied tags that describe their appearance. Keep in mind that characters that aren't very popular or don't have many images in FluffyRock's dataset typically won't fare that well without a LORA.
- Avoid weighting camera angle keywords too strongly, especially close-up.
- Resolutions between 576 and 1088 should work reasonably well as that is the range of FluffyRock.
A example for a solo focus pic:
feral pony fluttershy ( doing something / action ), equine, quadruped, pegasus, blue green eyes,
(feral:1.1) <- (this is weighted, using ( ) and :0.1-2.0 it makes a tag weaker / stronger for instance) friendship is magic, my little pony, meadow, ( a bunch of quality tags that change depending on what you want to go for )
BREAK ( separator that can help for multicharacter / after initial style prompt )
( other character, in this case offscreen human )
For multi character interactions it might be best to stablish style / number of characters / the type / perspective of the scene first then use a BREAK between the different characters. It helps it from getting them mixed up.
Use these embeddings:
https://civitai.com/models/87781/boringe621-negative-embedding-enhance-images-stylistically-and-topically
https://civitai.com/models/127533/fluffyrock-quality-tags
And this lora:
https://civitai.com/models/82098/add-more-details-detail-enhancer-tweaker-lora
And this lora for darker / nighttime images / better shadows:
https://civitai.com/models/13941/epinoiseoffset
To start learning prompting I would recommend to visit the furry diffusion discord and just browsing E621 and memorizing tags that go with the kind of pic you would want to make. I recommend checking these discords for tons of examples:
https://discord.gg/furrydiffusion
https://discord.gg/SQVcWVbqKx
SD install guide:
https://rentry.org/voldy
Model:
https://huggingface.co/zatochu/EasyFluff/tree/main
Grab these from hugging face and put them in the automatic\models\Stable-diffusion folder:
EasyFluffV11.2.safetensors
EasyFluffV11.2.yaml
And download this vae: https://civitai.com/models/23906/kl-f8-anime2-vae
And put it in the automatic\models\Vae folder.
After you open SD in your browser go to the extensions tab and download this extension with this link: https://github.com/Seshelle/CFG_Rescale_webui then restart it.
- Natural language has a strong impact on this model as of V7. It won't understand everything but it can get you places. And then when you can't squeeze any more out of it, you can use e621 tags to help refine your prompt. (or you can just use tags, either way will work.)
- Start your prompt with a short natural language prompt describing what you want, then pad it with e621 tags to refine specific concepts. As this is a v-prediction model, prompt interpretation can be a bit more literal.
- Many flavor words and artists from SD 1.5 work again.
- PolyFur is trained on MiniGPT-4 captions, so try being really flowery with your prompts and use sentences even. (check the generation info on the preview images to see how)
- If an established character isn't coming out accurately, try increasing the strength of their token and adding a few implied tags that describe their appearance. Keep in mind that characters that aren't very popular or don't have many images in FluffyRock's dataset typically won't fare that well without a LORA.
- Avoid weighting camera angle keywords too strongly, especially close-up.
- Resolutions between 576 and 1088 should work reasonably well as that is the range of FluffyRock.
A example for a solo focus pic:
feral pony fluttershy ( doing something / action ), equine, quadruped, pegasus, blue green eyes,
(feral:1.1) <- (this is weighted, using ( ) and :0.1-2.0 it makes a tag weaker / stronger for instance) friendship is magic, my little pony, meadow, ( a bunch of quality tags that change depending on what you want to go for )
BREAK ( separator that can help for multicharacter / after initial style prompt )
( other character, in this case offscreen human )
For multi character interactions it might be best to stablish style / number of characters / the type / perspective of the scene first then use a BREAK between the different characters. It helps it from getting them mixed up.
Use these embeddings:
https://civitai.com/models/87781/boringe621-negative-embedding-enhance-images-stylistically-and-topically
https://civitai.com/models/127533/fluffyrock-quality-tags
And this lora:
https://civitai.com/models/82098/add-more-details-detail-enhancer-tweaker-lora
And this lora for darker / nighttime images / better shadows:
https://civitai.com/models/13941/epinoiseoffset
To start learning prompting I would recommend to visit the furry diffusion discord and just browsing E621 and memorizing tags that go with the kind of pic you would want to make. I recommend checking these discords for tons of examples:
https://discord.gg/furrydiffusion
https://discord.gg/SQVcWVbqKx
not provided yet
Comments
0 comments posted