Learning more about LLM is next on my to do list.
I would follo the course from OPEN AI and LangChain
Run open-source models locally on your computer like `ollama run mistral` and it will be serving over localhost. Your app can then use any LLM locally by HTTP request or you can use an npm wrapper (npm i ollama) and it's as simple as `ollama.chat()`.
Libraries like LangChain (LlamaIndex, etc.) also come with the Ollama wrapper in the library. You can just `import { Ollama } from "@langchain/community/llms/ollama"`
(not affiliated with any of these, just used them as a dev)
Make sure to share on Show HN what you build :D