We are still actively working on the spam issue.
Difference between revisions of "Stable Diffusion"
(→Prompt Design) |
|||
| Line 5: | Line 5: | ||
== How to Use == | == How to Use == | ||
| − | * [https://beta.dreamstudio.ai | + | * [https://beta.dreamstudio.ai beta.dreamstudio.ai]: official web service |
* [https://github.com/CompVis/stable-diffusion Official Github page] | * [https://github.com/CompVis/stable-diffusion Official Github page] | ||
| − | * [https://github.com/basujindal/stable-diffusion | + | * [https://github.com/basujindal/stable-diffusion basujindal fork]: fork that uses less VRAM at the cost of speed |
| − | * [https://github.com/harubaru/waifu-diffusion | + | * [https://github.com/harubaru/waifu-diffusion waifu-diffusion fork]: fork that ??? |
== Example Prompts == | == Example Prompts == | ||
| Line 30: | Line 30: | ||
== Keywords == | == Keywords == | ||
| + | |||
| + | == Useful Links == | ||
| + | * [https://www.krea.ai/ krea.ai]: Website that lets you explore keywords | ||
| + | * [https://github.com/rom1504/clip-retrieval clip-retrieval]: Project that lets you determine the relationship between images and keywords, works in either direction. Online version [https://rom1504.github.io/clip-retrieval/ here] | ||
Revision as of 08:52, 22 August 2022
Stable Diffusion is an open-source diffusion model for generating images from textual descriptions. Note: as of writing there is rapid development both on the software and user side. Take everything you read here with a grain of salt.
Contents
How to Use
- beta.dreamstudio.ai: official web service
- Official Github page
- basujindal fork: fork that uses less VRAM at the cost of speed
- waifu-diffusion fork: fork that ???
Example Prompts
Prompt Design
Guidelines for creating better prompts.
Prompt Length
Be descriptive. The model does better if you give it longer, more detailed descriptions of what you want. Use redundant descriptions for parts of the prompt that you care about.
Note however, that there is a hard limit regarding the length of prompts. Everything after a certain point - 75 or 76 CLIP tokens depending on how you count - is simply cut off. As a consequence it is preferable to use keywords that describe what you want concisely and to avoid keywords that are unrelated to the image you want.
Punctuation
Use it. Separating keywords by commas, periods, or even null characters ("\0") improves image quality. It's not yet clear which type of punctuation or which combination works best.
Keywords
Useful Links
- krea.ai: Website that lets you explore keywords
- clip-retrieval: Project that lets you determine the relationship between images and keywords, works in either direction. Online version here
