34a78ae76d7240dab8f2004ece881306Q: Do you have video/tutorial/usecases/installation guide about "custom ai" ?
I plan to buy tier 3 .
overment
Jul 17, 2024A: Here's a template you can use to start or grasp the idea:
https://github.com/iceener/heyalice-nodejs-backend-template
Run this app on localhost:3005. In Settings -> Custom AI, set:
- Server URL: localhost:3005/api/chat
- Stream response: Active
- Local Model: Any name (e.g., "Alice")
And then you can select the "Alice" model from the list of available models and start chatting with your custom AI.
It's experimental right now but seems to be our most important feature. I'm working on image and file support, as well as syncing all settings and configurations with your custom AI. Chat support is now available. The app sends you the entire conversation, allowing you to implement RAG, conduct online searches, or create an agent-based workflow as needed.
Feedback is very welcome :)
Guys. I've purchased T3 but struggling to find any real value that I can't get from lots of other gpt assistants out there. If you can clearly establish how one can use the custom AI feature for in house knowledge base, rag, webhooks use cases then we cna start to see potential use cases that are fairly reasonable quick wins. I really don't want to spend a lot of time figuring out the simple stuff