An opinionated guide to which AI to use to do stuff
The definitive map: when to reach for Claude vs ChatGPT vs Gemini vs local models.

Why it matters
A practitioner's guide to model selection across real use cases—which models excel at coding, writing, reasoning, and local deployment—cuts through the noise of vendor claims and benchmark theater.
The key facts
9 to knowOpinionated comparison across major hosted models (Claude, ChatGPT, Gemini)
Use-case-driven recommendations: coding, writing, reasoning, local inference
Author: Simon Willison (prominent AI practitioner and commentator)
Published July 2026 (recent, presumably reflects current model capabilities)
Practical guidance on model selection rather than feature announcements
Published by Simon Willison (Datasette creator, noted AI practitioner)
Framed as practical guidance on tool selection, not marketing
Covers model/product selection by task type
Addresses the practitioner pain point of navigating competing options
Go to the source
Simon Willisonsimonwillison.net
