#12280 - explicit, ai generated, automatically imported, derpibooru import, generator:easyfluff v11.2, prompter:ada, princess ember, dragon, g4, cave, cloaca, dragoness, female, gemstones, genital slit, gold, horns, looking at you, lying, lying down, nudity, on side, solo, solo female, spread wings, tail, treasure, vulva, weird lizzer hole, wings

Image imported from Derpibooru.
Original: https://derpibooru.org/images/3264837
Metadata at the time of import:

Field	Data
Favorites	203
Upvotes	269
Downvotes	18
Score	251
Comments	0
Uploader	Ada

Original description below this line

About time I gave Ember a go.

Here’s more that didn’t make the cut if anyone wants:
https://ada321.itch.io/the-dragons-treasure

Figured I’d upload the model mix I use. It’s easyfluff 11.2 merged with some loras including 2 I made myself. I had posted it on 4chan some time ago as well:
https://drive.google.com/file/d/1e9x01E5k1NJfrh4xFWusDCglBWkltTt2/view

SD install guide:
https://rentry.org/voldy

Put both the safetensors and the yaml in the automatic\models\Stable-diffusion 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.
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.

I recommend checking these discords for tons of examples / prompting help:
https://discord.gg/furrydiffusion
https://discord.gg/SQVcWVbqKx

Syntax quick reference: **bold** *italic* ||hide text|| `code` __underline__ ~~strike~~ ^{^sup^} _~sub~

Detailed syntax guide

Image imported from Derpibooru.
  Original: https://derpibooru.org/images/3264837
  Metadata at the time of import:

Field | Data
  --- | ---
  Favorites | 203
  Upvotes | 269
  Downvotes | 18
  Score | 251
  Comments | 0
  Uploader | Ada

Original description below this line

---

About time I gave Ember a go.

Here's more that didn't make the cut if anyone wants:
https://ada321.itch.io/the-dragons-treasure

Figured I'd upload the model mix I use. It's easyfluff 11.2 merged with some loras including 2 I made myself. I had posted it on 4chan some time ago as well:
https://drive.google.com/file/d/1e9x01E5k1NJfrh4xFWusDCglBWkltTt2/view

__SD install guide:__
https://rentry.org/voldy

Put both the safetensors and the yaml in the automatic\models\Stable-diffusion 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.

- 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.

I recommend checking these discords for tons of examples / prompting help:
https://discord.gg/furrydiffusion
https://discord.gg/SQVcWVbqKx

https://derpibooru.org/images/3264837

Tantabus

BBCode

Comments