Somewhat reassured, I used the fully HA-integrated Claude Code to finish migrating most of my nearly 200 smart home devices.
In tests, Gemini 3.5 outpaced GPT-5.1 High and Opus 4.5 on coding tasks, giving you cleaner outputs and fewer fix-it loops.
Four chatbots agree on which team will win. I also got predictions on the coin toss, the halftime show, the most expensive ads, the national anthem and viewership numbers.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results