3 ms·

Yes, the repository has already been released, the weights will be released soon. And no, the content policy is nowhere near as strict as OpenAI's.

• yreg 4y ago
  I don't see the content policies[0][1] as much different, but perhaps it's a matter of opinion regarding the individual points. I don't know how strictly the policies are enforced, so perhaps there is a difference there as well. What I find strange is that StabilityAI doesn't outright ban using actual living people in prompts (where I can imagine wide usecases for abuse). At the same time both OpenAI and StabilityAI block sexual content/nudity/violence. I don't see why such content is supposed to be harmful. Anyway looking forward to the model, thanks for informing me about this project. [0] - https://stability.ai/stablediffusion-terms-of-service https://stability.ai/stablediffusion-terms-of-service [1] - https://labs.openai.com/policies/content-policy https://labs.openai.com/policies/content-policy
  • GaggiX 4y ago
    OpenAI with violence means that you cannot generate weapons (even knives), StabilityAI with violence means anti-Semitism, misogyny, etc. No one will mute you in generating nudes if it was not your intention, this means that the model is able to do this while the OpenAI model cannot. And anyway we are only talking about the limitation on the Discord server, when the model is free and open source you will be able to generate whatever you want.
    • yreg 4y ago
      I see, thanks.