Update 1.02 is now available.

Hello everyone, Jetro here.

Version 1.02 is now available. This new version of Janis AI features some much needed stability fixes, UI improvements, and the ability to use your own llamacpp compatible models.

Here's a list of changes in version 1.02:

• Improved backend that should deal with crashes some users were experiencing without the work-around used in the previous patch.
• Users can now use their own AI models! Janis AI now uses llamaCPP which means it can now support any model in the .gguf format using the ChatML template. You can find popular .gguf AI chat models on sites like huggingface.com.
• CPU mode. Users can now select a CPU mode to use with their AI model, which may be useful if you have a custom model that's too large for your GPU or if you don't have a minimum spec GPU.
• General improvements to the UI and quality of life fixes to make the user experience better.
• Added a text crawl for when your AI is replying similar to the text crawl seen in ChatGPT replies.
• Added a "Stop" button that allows users to interrupt the AI replies mid crawl.
• Added a User Name option in the Settings menu. Users can now input their name and your AI will remember it between all of your conversations.
• Improved the layout of the Create Persona UI.
• Added the ability to set a custom context length for the AI's memory.