Skip to content

Learn

Plain-language explanations of the terms and numbers you see across this site.

Guides

6 min readHow to choose an AI modelA practical, five-question method to pick the right AI model for your task and budget, without relying on hype or leaderboards.5 min readAI API pricing explained: tokens, input, output and cachingHow AI model pricing works, why output costs more than input, and how to estimate your monthly bill with a worked example.5 min readOpen-weight vs closed AI models: which should you use?The real differences between open-weight models like Llama, Qwen or DeepSeek and closed models like Claude, GPT or Gemini: cost, privacy, quality and effort.6 min readHow to cut your AI API bill: nine things that workMost AI bills are bigger than they need to be. Nine practical ways to pay less, from caching and batch discounts to sending the easy requests to a cheaper model.5 min readHow to test an AI model on your own taskLeaderboards tell you which models are strong in general, not which one is best for your job. A simple test you can run in an afternoon, with 30 real examples.6 min readHow to run an AI model on your own computerRun a capable AI model at home, free and offline, in about fifteen minutes: what your computer needs, LM Studio or Ollama, and which model size to pick.