• Murdoc
    link
    fedilink
    English
    arrow-up
    1
    ·
    1 year ago

    Just spent an hour trying to get sd to work, I guess it doesn’t like my system or something. :(

    • Ender of Games
      link
      fedilink
      English
      arrow-up
      3
      ·
      1 year ago

      Stable Diffusion does require some setup and knowledge, for occasional hobbyist use it’s probably too much to ask for.

      Midjourney sadly stopped free trials. But DALL-E and ChatGPT (OpenAI) have research/free use available. And DALL-E 3 costs money through OpenAI, but if you just search for “Bing image generator” it uses DALL-E 3. There are other free generation tools, and some examples on using SD and branches of SD online if you get interested in seeing who can do what. Sadly most free tools have censorship rules, but hopefully “hung” won’t trip any of them.

      • Murdoc
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 year ago

        Thanks! Know anything about InvokeAI? I’ve managed to get that to work, so it’s just a matter of finding a good model for it to use. And getting better with prompts of course, but communities like this are helping a lot there.

        • Ender of Games
          link
          fedilink
          English
          arrow-up
          1
          ·
          1 year ago

          I haven’t tried Invoke, no. My SD journey is at a standstill because I haven’t learned how to convert safetensors into working model files, and I haven’t tried in a few months.

    • woobie@lemmy.world
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 year ago

      Might be easiest to use the Bing Image Generator which uses Dall-E-3 as the backend. It’s currently free.