sorted by: new top controversial old
Mature-flag reviews should say what triggered — what got your flag lifted? in c/perchance@lemmy.world
[–] 3 points 1 week ago* (last edited 1 week ago) (1 child)

I agree a 'this is why' would be nice when your generator is tagged as mature, but on the same note - if the generator is or could create mature content, why put effort into having the flag removed if the cost is stripping out features? If you would prefer to make the changes and your intended audience is in fact only PG-13 or whatever the threshold is, then working to resolve the issues may be worthwhile - but if it is or could be used with mature content, then that is what it is, no reason to change it, but then it is also prudent to have the mature content gate and not a punishment.

  • permalink
  • fedilink
  • source
    •
    0
    Extending the Perchance AI Chat (perchance.org)
     

    While mostly a personal project, I decided to build tools to support and extend the 'official' ai-character-chat rather than fork it. I have seen several forks and thought achieveing many of the added features could be done by enhancing the perchance chat rather than replacing it. My Character Workshop does that and more, allowing a method of extending what the ai-character-chat does without forking.

    Certainly a work in process, the tools range from a utility to trim the fat off long chats, read them offline, build personal and public character and 'world' libraries, character generator, import and upgrade tools, image tools, and a World Builder that provides AI feedback and more.

    The landing page title the post grabbed is a bit confusing, it was intially forked from a character generator, but now ties together a series of generators and tools: https://perchance.org/ess-adv-char-gen

    How to get useful work out of the new AI Helper — prompting basics for gen builders in c/perchance@lemmy.world
    [–] 0 points 1 month ago

    While I agree that the latest AI Helper change was not an 'upgrade', I also agree that with some adjustment, it still works fine. And when compared to any linked / API driven AI model integrated into your IDE, there is no way any of us should be 'complaining'. If I would use GitHub Copilot at the same rate you can with the AI Helper, you could blow through your monthly token allotment (at GitHub Copilot Pro+ level) in a day or two, so we should be glad about what is being offered and adjust as Daveed mentions.

  • permalink
  • fedilink
  • source
    •
    An idea for saving cost. Local AI model use. in c/perchance@lemmy.world
    [–] 3 points 1 month ago* (last edited 1 month ago) (1 child)

    I would guess like many things, the perchance dev(s) have too much on the plate already and can only deal with the most pressing issues. That idea sounds good - a bridge for vscode or something that would allow outside programming assistance (which is probably somewhat viable through copy and paste now, though it would be clunky). The issue is just that though, I suggest vscode because that is what I use and you can implement all the popular AI assistants in vscode - but that is just one of many ways you could implement outside resources, and providing those code mods and plugins is a project in it's own.

    I assume perchance is avoiding a donation system, but the reality is it may be a best case - have the funds needed to be able to prioritize projects by another factor than needing to trim expenses, which are unlikely to go down (not much ever does).

  • permalink
  • fedilink
  • source
    •
    [dev] Short note in c/perchance@lemmy.world
    Unable to save changes to a generator (see screenshot) in c/perchance@lemmy.world