Let me try to put this in the scale of coffee: -- Using LLM via api: Starbucks. Inference at home: Nespresso capsules. Fine-tune a small model at home: Owni

Let me try to put this in the scale of coffee: -- Using LLM via api: Starbucks. Inference at home: Nespresso capsules. Fine-tune a small model at home: Owning a grinder and an italian espresso machine. Pre-training a model: Owning a moderate coffee plantation.