Welcome to Eudaimonia | Share, discuss, create

KairosJS

This is the heart of the community, everything about Eudaimonia lives here.

What belongs here

🎭 Scenarios/Personas — Share your creations, discuss narrative ideas, ask for help or feedback.
⚡ Stim configurations — E-stim (DG Labs, 2B...), Intiface/Buttplug.io - Share what works for you (always include a safety disclaimer — see Community Rules).
🖼️ LLM & Images & TTS — Tips on RP with AI, image generation, voice settings, immersion tweaks.
💡 Feature requests & ideas — What would you love to see in Eudaimonia?
📖 Tips & guides — Help fellow users get the most out of the platform.

Guidelines

When sharing stim settings, always mention your device model, your experience level, and add:

⚠️ These settings work for me. Start low and adjust to your own comfort. Never use settings you haven't tested gradually.

Be constructive when giving feedback on someone's creation. They put effort into it.

Welwa

Hi, thanks for creating this space, great idea!

Mesh1401

Hi Kairos, many thanks for creating all this.
And a belated special thanks for the option to play sound on vids, its great!
Since you've opened a forum, i'd like to grasp the opportunity to ask you a not so important question but one that has been puzzling me since the beginning. The question is, what on earth do you mean by "Outfit Top's description" and "Outfit Bottom's description" in the character generator, and what does it do? Is 'top' upper half of the body and 'bottom' lower half? Or is it more like outer clothes vs. underwear? Or is it perhaps more like dominant/submissive?
I've tried out different ways to fill it in but noticed no difference. But there's probably a reason it's there.
As said, not urgent but I hope you can finally clear up this mystery :-)
Thank you!

Welwa

Mesh1401
Hey Mesh, if you check the 6 characters that are provided, and click on them, you will see their data in those fields, and they are as you have guessed, the clothing of the upper body (top) and lower body (bottom)

KairosJS

Mesh1401 My broken English is probably confusing. In this case, "Outfit Top" refers to the upper half of the body, and "Outfit Bottom" refers to the lower half of the body. I will change the formulation of the labels.
It allows the system, in some cases (e.g., a close-up on the face), to remove the lower half from the prompt, as it's not needed and is likely to create some monstrosities.
By the way, I would recommend not adding any footwear tags, as the image model will often try to include them in the final image regardless.

Mesh1401

Welwa Aha, I see. Allright. Thanks!!

Mesh1401

KairosJS Thank you, that's perfectly clear. My confusion probably has to do with how these things are called in my native language as well.
Could be useful to put your explanation under a [?] button instead of changing the labels. Anyway, no big deal.

Snowcat22

I'm curious about a potential feature. I'm not sure how easy or difficult it would be, or if it is even possible with LLMs. Is there a way for the AI to bring two different characters at the same time? Envision two mistresses with control of the stim box, perhaps each with different schemes. Or imagine the recent 'birthday party' scenario, which I really enjoyed (and lost) btw, but your girlfriend gives the remote to Sarah or Jessica for a bit.

Anyway, I think this is would be a cool feature, if the AI can handle it.

KairosJS

Well, multi-character is still hard for both LLM and image generation models, but it's way better today than it was a year ago, especially with very large models such as Claude.
That said, Eudaimonia hasn't been developed with this in mind, so for now, this is something I'm setting aside. But I hope we'll be able to do it in reliable ways in the future.

Snowcat22

KairosJS
I get that it would be difficult for image generation, but I don't use or have much interest in that, only the LLM. Thank you for your hard work and making this space!

MalteP

KairosJS
What about censored image generation ? Basically giving the LLM the option to further punish the user visually visually by censoring the generated image. I imagine this should be quite doable with either additional layers in comfyui after the image creation or/in combination with Model like these: https://github.com/vladmandic/nudenet, https://github.com/silveredgold/beta-censoring

What I would also find quite interesting, would be the ability to add custom MCPs to the model to further enhance it. For example I would like to give my domme the ability to search pornhub. I could start an mcp server using this pornhub api: https://github.com/Derfirm/pornhub-api and through this the llm could actually send actual video links for the user to watch knowing their descrition and tags enabling it to comment a lot more accurate what the user is watching.

KairosJS

MalteP Tahank you for the resources!
I've started working on a censor module. As I was planning to add MediaPipe for different purposes, I think I will use it instead of NudeNet. I need to test whether I can get consistent results with MediaPipe.

Custom MCPs can be interesting, I will think about how to implement this.