🔴Create A Local AI Voice Assistant With A Customizable Persona 👉 ruclips.net/video/PltEIx3LOLg/видео.html 👉 Want to reach out? Join my Discord by clicking here - discord.gg/5hmB4N4JFc
Well the thing is, you can use local llm models through projects like ollama, lm studio or any local provider with a compatible OpenAI API, with the web search feature if you like but from my testing, models smaller then 70B are not that great at using the tools. I was using the groq api in this video with the llama 70B model which can function call and work with these kinds of tools better.
@@TheLocalLabIf code export option is there, then it will be easy for non coders to use the code for similar usecase. If export option is there, then we can export and use the same code to complete the task in short period if time
Excelent video. It is very useful. Thanks for sharing it!!!
🔴Create A Local AI Voice Assistant With A Customizable Persona 👉 ruclips.net/video/PltEIx3LOLg/видео.html
👉 Want to reach out? Join my Discord by clicking here - discord.gg/5hmB4N4JFc
can you search the web using a local LLM it seems like your not using a local hosted LLM to search the web
Well the thing is, you can use local llm models through projects like ollama, lm studio or any local provider with a compatible OpenAI API, with the web search feature if you like but from my testing, models smaller then 70B are not that great at using the tools. I was using the groq api in this video with the llama 70B model which can function call and work with these kinds of tools better.
can you show us about the SQL Connector, mine never success doing it, like the AI can not connect to the DB.
Using gpt-4o for successfully working
Thank you for brief overview! Does it support image input?
Yes it does. As long as your using a vision model as your llm.
Is it possible to export the code?@@TheLocalLab
@@KalaivaniAruchamy Code for what exactly?
@@TheLocalLabIf code export option is there, then it will be easy for non coders to use the code for similar usecase. If export option is there, then we can export and use the same code to complete the task in short period if time