Same SD mistake as always with waterfalls. The small ones in the front are captured with a long shutter time. The ones in the back show water standing still. It's a nice creation, but clearly AI generated.
We could start sharing some kind of content. Like chats or stories. Then including generation parameters would be something "unique" and may make this sub a place to which people return. IMHO this platform is way more suitable for that than discord. Alternatively a character mega treat. Probably separate sfw and nsfw then. What are your ideas?
Will Verstappen be p1 before DRS gets enabled? Place your bets. 👇
Ja, en dan wat specifiek een zachte L
I don't totally understand your question. But on relatively slow hardware I would think to focus on ways of creating multiple images and variations of prompts while you are absent would be a bonus.
https://lemmy.world/comment/1595398 summarizes it nicely, altough without references.
Duplicati to Backblaze B2 for the important stuff. For as far as the media library goes, no backup just local raid setup...
When I send a message to the AI, it'll often generate the character's response as well as my response to what they said, in the form of "You: ..." I assume this has something to do with the formatting of the sample dialogue, or with the generation parameters. Does anyone know how to fix this?
In theory the current models may by accident get it right. However since there is no detailed information available about what different rooms are required, their function their size etc. The answer is no, not yet at least.