Welcome to Yiff AI
Yiff AI is an AI image and video generator powered by furry-focused Stable Diffusion models, including our own Yiff AI Advanced V0.8. Whether you're a first-timer or an experienced prompter, this guide will help you get the most out of the platform.
Write your prompt in English
The models understand English e621 tags. A prompt in another language gives poor results. Not comfortable in English? Type your idea in any language and click Enhance prompt: it turns your description into English tags for you.
How It Works
The generator takes your text description (called a prompt) and transforms it into an image. Each word in your prompt carries importance and power - the AI reads your tags and builds the image based on them.
Your First Image
- Go to the Generate page
- Select a model (we recommend Yiff AI Advanced V0.8 for the best results, see Models)
- Type your prompt in English, using tags separated by commas - for example:
solo, anthro, male wolf, grey fur, muscular, sitting, forest background. A prompt needs at least 40 characters. - Choose your image size (
1024 x 1024is a safe default) - Click Generate
Tags, Not Sentences
The AI works best with tags separated by commas, not full sentences. Think of it like a list of keywords describing what you want to see.
| Less effective | More effective |
|---|---|
A muscular male wolf with blue eyes sitting in a forest | male wolf, blue eyes, muscular, sitting, forest background |
Tag Order Matters
Tags placed earlier in your prompt have more influence on the final image. Put the most important elements first - typically the character count, species, artist style, and key features.
Recommended Prompt Structure
For the best results, follow this order:
- Character count -
solo,duo - Artist style -
by Chunie,(by meesh:1.3) - Species & gender -
anthro, male wolf - Physical traits -
grey fur, muscular, blue eyes - Pose & action -
sitting, looking at viewer - Background -
forest background,bedroom - Rating -
safe,explicit
Reusing & Iterating
Every image has actions that send it back into the generator, so you can build on what you already made:
- Reuse prompt - reloads the generator with that image's prompt and negative prompt, ready to tweak and generate again.
- Reference - opens the generator in img2img mode with that image set as the reference, so your next generation stays close to it. See Creativity Slider (img2img) to control how far it can deviate.
- Generate video - opens the generator in Video mode with that image as the start. See Video Prompts.
- Inpaint and Upscale - edit part of the image or enlarge it, on Yiff Master and above. See Inpainting and Upscale.
You'll find these on any image, both in your gallery and on the image's own page. Before you start, take a minute to read the Content Rules.