It's possible to use a local LLM in place of OpenAI. My personal experience has been that using Ollama is easiest. The question of which model to use will quickly become outdated and depend on the hardware that they're running.
I've also used LMStudio, but that requires users to jump through more hoops that Ollama. If they're not technical, it can be a pain to support.
It's possible to use a local LLM in place of OpenAI. My personal experience has been that using Ollama is easiest. The question of which model to use will quickly become outdated and depend on the hardware that they're running.
I've also used LMStudio, but that requires users to jump through more hoops that Ollama. If they're not technical, it can be a pain to support.