I’ve heard training on synthetic data is fine now. Most datasets are augmented with synthetic data.
I think it’s that way on purpose.
Tiny Alligator-turtle hybrid, on rock by steam, sunset, reeds, flowers, rim lighting
Steps: 8, CFG scale: 1, Sampler: Euler, Seed: 367283036, Size: 832x1216, Created Date: 2024-07-09T0536:23.6892411Z, Clip skip: 2
I forgot what passes for dining in some parts of the country.
The table isn’t set for breakfast yet.
Comfy gets a lot of stuff sooner than A1111 these days, but I only use A1111, so I don’t really know.
This whole license is revocable, and they still want to force you to delete your model and All derivative work if they say you invalidated the license for any reason.
Reminder that the author of Glaze, Ben Zhao, a University of Chicago professor stole open source code to make a closed source tool that only targets open source models. Glaze never even worked on Microsoft, Midjourney, or OpenAI’s models.
It’s about 20 episodes too long and you really feel it.
It’s so bad, but the ending is worth looking up.
So Kadokawa had this failing mobile game named Kemono Friends, but they wanted a TV anime for it, so they get Omoto Tatsuki to direct it. He takes their product and actually turns it into something people want to watch, going as far as, releasing supplemental clips from the show on Twitter and even arranging a collaboration with the Tobu Zoo in Saitama Japan. This collab gave us the tragic saga of Grape-kun the Humbolt Penguin, a meme in its own right. Since the show didn’t have much of a budget he hired no-name voice talent, launching their carriers thanks to the popularity of the show.
Ultimately, it was the clips on Twitter that did him in if rumors are to be believed. He was sharing clips as usual, but Kadokawa wanted that content for the home video special features, so they fired him abruptly. He already saved Kemono Friends for them, so they cut him loose.
I’m interested to see who else they screwed over, considering they also got caught bribing the Tokyo 2020 Organizing Committee to secure a sponsorship deal.
I’m not a bot.
My first time seeing someone go to bat for Kadokawa, considering how they treat their employees.
Kadokawa group sucks.
I hope they release it.
Someone needs to fine-tune a model on that translation.
It’s pretty cool. https://github.com/Acly/krita-ai-diffusion?tab=readme-ov-file
Dunno, but you could ask the OP I linked in the text.
Here’s one. Did they overfit their model and think they could block the bad prompts?
You’re always at the top of the most reacted.