For prompts ideas you also can browse AI images from deviant. Some posters there (like
Regarding the negative prompt, it can help for sure but do not abuse of it or you'll get always the same stuff. sometimes a better described positive prompt is more efficient.
Ex:
- Remove "ugly" from the negative prompt to avoid to get always barbie's looking girls but be more descriptive in your descriptive prompt (pretty, cute are working well, but elegant, natural, next door also), you'll get more realistic women.
- cartoon, 3d in the negative is plenty enough to get photo like results
Personally I'm using the Liberty model and I subscribed to the patron of its creator, he provides there a very insteresting prompting guide. Do not ask, I won't share it in the respect of this creator.
I'm also using the Deliberate model which provides very nice results.
I don't like too much URPM as I'm more into erotic.
And you also can merge models together with the Ckeckpoint merger tab.
For tag weight explanation (from
You must be registered to see the links
) forget to remove their metadata, so just upload picture from Deviant in the PNG info tab to get the prompt and parameters.Regarding the negative prompt, it can help for sure but do not abuse of it or you'll get always the same stuff. sometimes a better described positive prompt is more efficient.
Ex:
- Remove "ugly" from the negative prompt to avoid to get always barbie's looking girls but be more descriptive in your descriptive prompt (pretty, cute are working well, but elegant, natural, next door also), you'll get more realistic women.
- cartoon, 3d in the negative is plenty enough to get photo like results
Personally I'm using the Liberty model and I subscribed to the patron of its creator, he provides there a very insteresting prompting guide. Do not ask, I won't share it in the respect of this creator.
I'm also using the Deliberate model which provides very nice results.
I don't like too much URPM as I'm more into erotic.
And you also can merge models together with the Ckeckpoint merger tab.
For tag weight explanation (from
You must be registered to see the links
)- a (word) - increase attention to word by a factor of 1.1
- a ((word)) - increase attention to word by a factor of 1.21 (= 1.1 * 1.1)
- a [word] - decrease attention to word by a factor of 1.1
- a (word:1.5) - increase attention to word by a factor of 1.5
- a (word:0.25) - decrease attention to word by a factor of 4 (= 1 / 0.25)
- a \(word\) - use literal () characters in prompt
