Let users choose between local/hosted inference & cloud APIs

[frgfm]

It makes sense that some users won't have compatible hardware to run LLMs locally. In that case, they might want to use external APIs for this.

It could be interesting to provide the following options:

• Ollama (feat(code): switch fully to ollama as LLM provider #101)
• OpenAI (feat(llm): add alternative support of OpenAI API #163)
• Mistral
• Claude
• Gemini
• Groq (feat(llm): enable the choice of LLM provider #157)

[bright258]

I agree.

[frgfm]

I agree.

@bright258 which LLM provider would you like to use most? We now have full support of Groq & Ollama

[bright258]

OpenAI

[frgfm]

OpenAI

@bright258 Done :) Just merged #163