March 25 - Estimai Online and Stereo box release notes
A new update has been deployed.
I want to thank all of you who subscribed to my patreon, your support is warmly welcomed and it makes me even more motivated. Thank you!
I also want to deeply thank JBK from Milovana who built the first set of stims for stereo boxes and made it possible for eudaimonia to support DIY boxes.
EstimAI Online
A big new feature for those who want to try EstimAI but can't use it locally. EstimAI is now available online for all Eudaimonia Patreon subscribers.
You can find all relevant informations here!
It features all the local options and will likely allow new possibilities in the future.
You may be interested in EstimAI Online if
- You want to try ERP with AI
- You can't use local AI
- You want to enjoy both: Text (online) & Image generation (local) at the same time
- You want faster, easier set up
- Use stronger models
- It's cheaper than using runpod
For those who don't want to use Patreon, I'm working on a Paypal & Stripe integration, which I will add as soon as possible.
Stereo box
Stereo box support has been added to eudaimonia.
- Uses its own set of stims, built by JBK for stereo boxes.
- Frequencies are limited from 500hz to 1500hz, and the maximum possible volume is -5dB.
- As with other devices, you can easily create and share your own stims and manage your collection.
Intelligent images set
A new image set type has been added, the first two of which were released with this update: Saeko and Ladria.
They have been created to match images with action better. This is still WIP as it's not fully integrated yet. Only a few scripts exploit the potential of these sets.
Also, it only works for normal eudaimonia sessions. Support for EstimAI is currently being studied, probably not coming soon (as it runs completely differently).
You can find Saeko's Lora on my civitai
I may publish eudaimonia's realistic characters Lora later, but they are of pretty poor quality to be honest.
Image generation
Small improvements have been made to image generation
- Thanks to a user who gave me the idea, a low-resolution option has been added to improve generation time for low-performance hardware.
- Pony realistic prompts have been improved a bit.
- Many prompts have been refined or removed, to reduce the nightmare's fuel.
A lot of new open source video models have been released recently and they look promising. Although I'm still studying the matter, it's possible that short, simple videos will be added to the images sets at a later date!
EstimAI Improvements
Small improvements have been done to EstimAI.
- The prompt has been slightly improved.
- Personas of standard scenarios have been improved.
I hope you like it!
Don't hesitate to contact me if you encounter any issues or want to share ideas and feedbacks.