Which Model To Use?

v1: Sandwing | v2: Blister

• Sandwing (V1) can be used to generate generic sandwings, and specific characters such as Sunny, Blaze, Thorn and Qibli

• Queen Coral (V2) can be used to generate Blister. I recommend combining this LoRA with the base LoRA to reduce distortions. (See notes on how I do this or check out my pictures)

Blister (V2)

Blister from the book series Wings of Fire. Created a LORA for her since her scale pattern is rather unique that I didn't include in version 1, but I realize doing a solo LoRA like this should be no problem now.

A mix of SFW and NSFW pictures were used for this model, with some help from the graphic novel as well too.

I would recommend combining both this LoRA and V1 at the same time. See notes on what I did.

It should be able to generate:

• With weightage about 0.8 to 1 for best results.

• Anthro and Feral (Put "Feral" or "Anthro" trigger words accordingly)

• SFW and NSFW
  

Prompts

Trigger Words: Blister_wof

Additional prompts: black scale pattern, black sclera, black eyes, female, 1girl, circlet

Optional prompts: Sandwing, dragon, yellow scales, horn, claws, wings, tail, tail barb

Notes

• Model is trained on the base Pony Diffusion model.
  

• Recommended to use the base version as well since the version is trained on a lot of sandwing pictures. Generating just V2 alone might cause more errors and distortions. My settings for V2 is set at 0.8, and V1 at 0.6 so it picks up more on Blister but still can generate the sandwing anatomy.
  

• Accuracy of the model is about 80 to 90%. Managed to generate her proper eyes by using 'black sclera, black eyes' combined with an eye mod to bring up the quality. See my pictures for an example.

• For anthro models, use "hands" and "fingers" in prompts for more proper human anatomy (standard 4 finger 1 thumb) or "claws" if you want a slightly more dragon (typically generates 4 fingers)

• For feral models, I tend to include these in negatives: "abs, fingers, belly button, hands, pov, human".

V1

Sandwings from the book series Wings of Fire. A mix of SFW and NSFW Images were used to train this model.

It should be able to generate:

• With weightage about 0.8 to 1 for best results.

• Anthro and Feral

• SFW and NSFW

• Male and Female models

• Characters such as Sunny, Thorn, Blaze and Qibli (Model was trained on most of their images.)

Prompts

Trigger words: Sandwing

Additional Prompts: yellow scales, dragon, scales, horns, claws, wings, tail

Optional Prompts: obsidian black eyes, tail barb

Prompts for Thorn: Thorn_wof, amulet, female, earring, brown hair

Prompts for Sunny: Sunny_wof, bright yellow scales, female, green eyes

Prompts for Qibli: Qibli_wof, earring, male, yellow hair

Prompts for Blaze: Blaze_wof, female, white hair

Notes

• Model is trained on the base Pony Diffusion model.

• "Tail barb" prompt can be an issue while generating. Add on the positive prompts "black tail barb" and "pointed tail barb" that could make it more accurate. I guess as long as it looks something dangerous, it should work. However barb generation can be still an issue in this version.
  

• You can use hair color to influence their hair color if the character generates a wrong one. For Sunny, I used "orange hair" for positive prompt for a more accurate character.
  

• For Sunny as well, you can include "black obsidian eyes" and "tail barb" in negatives and remove the "yellow scales" prompt. Add "orange wings" as well.
  

• Thorn is unable to generate her eye of onyx. If there's a v2, i'll try to train the model for the amulet itself if needed.
  

• For Sandwing eyes, it might be a hit or miss as well. Prompt is "obsidian black eyes" but may still generate normal black eyes. Adding "Black Sclera" to positive prompts may help.
  

• For anthro models, use "hands" and "fingers" in prompts for more proper human anatomy (standard 4 finger 1 thumb) or "claws" if you want a slightly more dragon (typically generates 4 fingers)

• For feral models, I tend to include these in negatives: "abs, fingers, belly button, hands, pov, human".

Want to support me without Tipping? Here's how:

1. Liking my posts, collecting my images or collecting my models all generate buzz for me.
   

2. Generate more pictures of my models. I get a base commission from anyone who generates using my model, even without the tip.

3. Leave a like on my model. This helps me with the algorithm in reaching more people

4. Leave a review. I take all feedback very seriously, good or bad.
   

5. Consider using the 'add post' under gallery... if you are posting pictures of just my LoRA. The images will also pictures will still appear on your profile but this will give me A LOT of buzz, while you can also complete your daily post quest.
   
   

   Thank you very much!