Hopefully you add some features because as is it's pretty bare bones. Most people would be better off with Silly Tavern or something. Not being able to have your own personality for the player character kills this for me. It's really just a text box, and a picture.
I have plans for many interesting features in future updates.
Elixir is designed to be easier to set up and use than Silly Tavern.
To use Silly Tavern you need to have a descent understanding of how AI Roleplay works to set the parameters, select the best LLM and adapt the character to run flawlessly on that LLM.
In Elixir, I have done all those steps for you and adapted the characters to run perfectly on fluffy/l3-8b-stheno-v3.2 (q4_k_m).
But, there's no way to "explicitly" describe "user Persona" in Elixir. The conversation summary system generates a summary of "user's personality" based on the conversation, and it becomes a part of the character's memory.
In short, how you behave with her (character) and what you tell her about yourself will define your persona for her. That's my current approach towards "user persona".
You are welcome, bro, but there's no way to toggle it off, because it's very important to keep the character's behaviour consistent, it tracks her attire, mood, feelings etc. and acts as her long term memory. I'll improve this system in upcoming months.
But even in its current state, the summarization takes very little time if you run ollama on Google colab with the colab ollama script included in the app.
The setup process is as simple as clicking a few buttons. Just check the short video guide in the setup instructions folder.
Elixir’s memory system is designed to be lightweight enough to run on low-end hardware, but its speed relies entirely on your Ollama server's processing power.
Running Ollama locally without a dedicated GPU (at least 8GB VRAM) will cause slowdowns. That’s exactly why the app includes a setup to run the Ollama server on Google Colab, taking advantage of their T4 GPUs for incredibly fast processing. Summarizing your entire chat history into the character's long-term memory within ~ 30 seconds.
This step is essential: since fluffy/l3-8b-stheno-v3.2 (q4_k_m) has an 8k context limit, we must compress past conversations to keep the story moving without breaking the context limit.
← Return to game
Comments
Log in with itch.io to leave a comment.
What is the difference between Lite and Paid?
Paid version comes with a lot of pics
Pics thar updates the character during certain triggers?
Not yet bro, but I am going to implement a gallery like system for future updates.
The pics are to visualise the character in various scenarios according to her stories.
If you want to change a character pic, let me know, I'll explain the process.
Sorry bro, not yet, may be in future
Hopefully you add some features because as is it's pretty bare bones. Most people would be better off with Silly Tavern or something. Not being able to have your own personality for the player character kills this for me. It's really just a text box, and a picture.
I have plans for many interesting features in future updates.
Elixir is designed to be easier to set up and use than Silly Tavern.
To use Silly Tavern you need to have a descent understanding of how AI Roleplay works to set the parameters, select the best LLM and adapt the character to run flawlessly on that LLM.
In Elixir, I have done all those steps for you and adapted the characters to run perfectly on fluffy/l3-8b-stheno-v3.2 (q4_k_m).
But, there's no way to "explicitly" describe "user Persona" in Elixir. The conversation summary system generates a summary of "user's personality" based on the conversation, and it becomes a part of the character's memory.
In short, how you behave with her (character) and what you tell her about yourself will define your persona for her. That's my current approach towards "user persona".
Hi, is there an option to toggle Memories, statuses and summaries of. After chatting for a while they take a long time. Thank you dev!
You are welcome, bro, but there's no way to toggle it off, because it's very important to keep the character's behaviour consistent, it tracks her attire, mood, feelings etc. and acts as her long term memory. I'll improve this system in upcoming months.
But even in its current state, the summarization takes very little time if you run ollama on Google colab with the colab ollama script included in the app.
The setup process is as simple as clicking a few buttons. Just check the short video guide in the setup instructions folder.
Have a nice day.
Detailed Explanation of Summarization Process:
Elixir’s memory system is designed to be lightweight enough to run on low-end hardware, but its speed relies entirely on your Ollama server's processing power.
Running Ollama locally without a dedicated GPU (at least 8GB VRAM) will cause slowdowns. That’s exactly why the app includes a setup to run the Ollama server on Google Colab, taking advantage of their T4 GPUs for incredibly fast processing. Summarizing your entire chat history into the character's long-term memory within ~ 30 seconds.
This step is essential: since fluffy/l3-8b-stheno-v3.2 (q4_k_m) has an 8k context limit, we must compress past conversations to keep the story moving without breaking the context limit.
How i can add more characters? Or replace for anothers
I haven't added a character editor yet, but actively working on such a system