Spaces:
Paused
Open-source appreciation thread
I just wanted to be the first community discussion on here, and to thank deepfloyd for making an image model better than midjourney. That is all I have to say,
FIRST!
I'm very impressed personally. It didn't quite get the text in my prompt right, but it's like infinitely better than anything else out there. Awesome job guys! Hope I pushed it to it's limits and got you some good data!
Giant, bright, blue neon sign that reads "The words of the prophets are written on the subway walls and tenement halls". Glowing in the night. Lighting up the surrounding landscape. Small shadows of people below. Realistic. Haze.
I'm incredibly excited to see where the community takes this model!
The license is kinda restrictive imo, feels openAI "open-source"
Yeah, but its nowhere near openai levels. it's just not quite as open as stable diffusion.
ngl im kinda stupid XD
fair, maybe more like Seek Art Mega License or Dreamlike but idk how high that bar is
Thanks for creating such cool model, the text ability is great. So far, I got issues on face after upscale. Face would be distorted and not clear after upscaling. Hope to hear your expertise.
For balance: Restrictive License Angry Thread
The license is kinda restrictive imo, feels openAI "open-source"
You understand that this is only restrictive as this initial release is just for experimentation and research purposes? They have stated that after the time goes by and the model keeps improving they will release it under true FLOSS licenses
They have stated that after the time goes by and the module matures enough they will release it under true FLOSS licenses
idk if rhat is meant to make them sound more or less like openAI. wouldn't mind a link tho
Sure! Here you go! https://twitter.com/stabilityai/status/1651984145944547329?s=46