What is GPT-5.5 Thinking, and when should I use it?

GPT-5.5 comes in two flavors that suit different jobs. The standard model is fast and conversational, ideal for the bulk of everyday use: writing, brainstorming, quick answers, and back-and-forth. GPT-5.5 Thinking is the reasoning mode, which deliberately spends more time and compute working through a problem step by step before responding, trading speed for accuracy on hard tasks.

You should reach for Thinking when the problem genuinely requires careful reasoning: multi-step math, logic puzzles, complex coding and debugging, dense analysis, and anything where a plausible-sounding wrong answer would be costly. The extra deliberation meaningfully reduces errors on exactly these kinds of tasks.

For most writing, summarizing, and casual questions, the standard model is the better choice because it is faster and the reasoning depth is unnecessary. Using Thinking for a simple email just makes you wait longer for the same result. Matching the mode to the difficulty of the task is the skill.

On Plus you get generous access to both, with the Thinking model capped at a large weekly allowance; heavier reasoning use is where the Pro tier's bigger quotas start to matter. For prompt techniques that get the most from reasoning models, see our ChatGPT prompts guide.