Ollama is still the easiest way to start local LLMs, but it's the worst way to keep running them
… A Mistral B Gemma C GPT-4o D Llama 3 Exactly right! GPT-4o is a proprietary OpenAI model only accessible via the OpenAI API — it cannot be run locally through Ollama. Ollama supports open-weight models like Llama 3, Mistral, Gemma, and many others. …