You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A question, perhaps, for @Aisuko: Is Kirin suitable as a chatbot for website visitor questions on open source project documentation, like https://docs.meshery.io? Or does this require a fair bit of model training and hosting effort?
Yes, we can do that. Our target is cost-effective AI solution for all the open-source projects. We use RAG technique and it has some benefits I mentioned in meshery/meshery#12777 (comment).
For the model training, I think we can investigate the it in the second step. For example, if we found that RAG is not enough for us. We can deign fine-tuning and evaluation pipeline for the model which means we do not need to spend any resources on the fine-tuning tasks.
There are two options,
1.self-host LLM(local ai) which needs cloud environment to host the LLM.
Contact Details(optional)
You can find me in https://slack.layer5.io, if you'd like to chat.
What feature are you requesting?
A question, perhaps, for @Aisuko: Is Kirin suitable as a chatbot for website visitor questions on open source project documentation, like https://docs.meshery.io? Or does this require a fair bit of model training and hosting effort?
More context: meshery/meshery#12777 // @vishalmvivek
The text was updated successfully, but these errors were encountered: