Why isn't this Ollama-centric?

[MikeyBeez]

Checked

• I searched existing ideas and did not find a similar one
• I added a very descriptive title
• I've clearly described the feature request and motivation for it

Feature request

We should be doing as much local processing as possible. Certainly we want to access foundation models when appropriate, but most processing should be done locally. That's just common sense. I know that OpenAI supports you. I know that they want blockware that forces people to use their LLM for everything, but it's harming the community.

Motivation

AI is too important to slow down its progress with blockware. No matter how much money you make from AI, that won't matter if a cure for what kills you comes one day late.

Proposal (If applicable)

Make your examples Ollama-centric. Llama 3 is amazing. Show people how to use foundation cloud services like OpenAI when appropriate.