ChatGPT, Claude, Gemini, Microsoft Copilot and Perplexity are all good general assistants by 2026; choose by the ecosystem you already work in, the jobs you do most, and the data terms you need, then try two side by side for a week.
Pick a coding assistant by where you work (terminal, IDE, or delegated cloud tasks), which models you want, and how billing works, then judge it on a real week of your own tasks rather than on benchmarks.
Choose where to call a frontier or open-weights model in production (first-party API, cloud platform or inference host) by models, tool use, caching, privacy and price.
To run open-weight models on your own machine, use Ollama for the easiest command-line start, LM Studio for a point-and-click app, llama.cpp for maximum control on modest hardware, and vLLM when you are serving many users from a GPU server.
Buy an off-the-shelf AI product when the job is common and not your differentiator, build on a model API when the workflow is yours, and only train or self-host models when data, scale or control force you to; decide with total cost of ownership and a 90-day pilot, not a demo.
Match the model tier to the task (small and fast for high-volume simple work, mid-tier for most products, the frontier tier for hard reasoning and long agent runs, open weights when data or cost demand it) and confirm the choice on your own eval set, not on leaderboards.