January 25 - Release notes

Here is the first update of eudaimonia, I wanted to thank the users who took time to give feedbacks, appreciations and cool ideas to develop.
First of all, I'm sorry to say that the Coyote3 version is still not ready. I've had some personal problems that have put me behind schedule. But that's my top priority for the next release.

Real-time image generation

This new feature will allow you to generate images of your favorite character from an AI during a session. A full guide on how to use it can be found here :
Guide

It uses ComfyUI, models from Pony's family and character LORA.

  • You can use the character editor to create up to 15 characters to play with.
  • You can choose between anime and realistic styles. However, I recommend that you use the anime style, which offers better-quality images.
  • This feature will make it possible in the future to display more adequate images of what's going on.

EstimAI

A lot of improvements have been made to EstimAI. I keep tinkering with the prompt, but I think this version is solid and rich. I hope it will make your experience even better.
With the release of Llama-4 just around the corner, and the models getting more and more powerful, I'm really optimistic about the future.
I'm also working on some cool features I've received from users to make the experience even more immersive.

  • Added more sampling settings especially DRY & XTC
  • You can now pick different estim system prompts to fit better the custom scenario you're playing
  • All standard scenarios have been reworked and added to the list of custom scenarios

The question of which model to choose is a frequent one, and there are more and more models to choose from, so I've created a benchmark page here.


Public Estims

You can now craft and share your own custom stims. For now it's still limited to a few categories, but eventually, it will be possible to have a fully personalized estim experience across eudaimonia.
Estims Page

This also opens up the possibility of building a fully DIY stereo box compatibility.

  • You can share/create stims with the community
  • You can manage your collection (stims playable during a session)

Dual channel

The whole estim's handling has been redesigned to use the two channels of your box (A channel dedicated to genitals, B to anus).

  • In EstimAI will the AI can now choose between channel A, B or both
  • I hadn't time to create content dedicated to exploit this feature, but that's coming later.
  • There is an option to still send the same signal to both channels for those who still want to use only one channel.

Miscs

And also a lot of small improvements (like in session message's position and size), bug fixes and a bit of new content.
I didn't have much time to create new contents (Images sets, new scripts for normal session, synced gifs, etc.) but I intend to add more later.


I hope you like it! I've still a lot of ideas to implement, and I will keep working hard on Eudaimonia and EstimAI. Don't hesitate to contact me if you encounter any issues or want to share feedbacks.