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.
Some results have been hidden because they may be inaccessible to you
Show inaccessible resultsSome results have been hidden because they may be inaccessible to you
Show inaccessible results