ChatGPT Instant now auto‑routes models

ChatGPT’s “Instant” button now means something different. It is no longer just the fast model. It is now a traffic cop that can quietly decide whether your prompt should stay on GPT-5.5 Instant or get bumped to GPT-5.5 Thinking instead. OpenAI also made GPT-5.5 Instant the new default model in ChatGPT, replacing GPT-5.3 Instant for logged-in users. ### What actually changed? The big shift is simple — “Instant” is now an auto-switching mode, not just a fixed backend. If you pick Instant, ChatGPT can decide on its own whether your request should run on GPT-5.5 Instant or GPT-5.5 Thinking. More complex prompts may get routed to Thinking for deeper reasoning before you see an answer. ### So is GPT-5.3 still in the picture? (help.openai.com) This is the confusing part. One Help Center page says Instant can auto-decide between GPT-5.3 Instant and GPT-5.5 Thinking. But the newer release note says GPT-5.5 Instant is now the default and replaces GPT-5.3 Instant for all ChatGPT users. The cleanest read is that OpenAI is in the middle of a documentation transition, and the live product direction is clearly toward GPT-5.5 as the default family. ### What does the user see? Usually, less than before. If you manually choose Thinking, ChatGPT will show a Thinking trace even for short reasoning runs. But if Instant silently routes your prompt to Thinking, ChatGPT may not always show that trace when the reasoning is brief. In other words, the system can use a slower, deeper model without making the switch very visible. (help.openai.com) ### Who gets manual control? Paid users get the real knobs. Plus, Pro, and Business users can use the model picker to manually select GPT-5.5 Instant or GPT-5.5 Thinking. They can also open Configure to turn automatic switching on or off, access legacy models, and set thinking effort when Thinking or Pro is selected. GPT-5.5 Pro sits above both, but only for Pro, Business, Enterprise, and Edu plans. (help.openai.com) ### What happens to older models? Older named models keep getting folded forward. OpenAI says GPT-5 Instant and GPT-5 Thinking were retired in ChatGPT on February 13, 2026. Older chats that used those models now continue on GPT-5.3 and GPT-5.5 equivalents, which means continuing an old conversation may not produce the same style or behavior you saw before. ### Why keep GPT-5.2 Thinking around? (help.openai.com) Because migrations break habits. OpenAI says GPT-5.2 Thinking will remain available in Legacy Models for 90 days after GPT-5.5 Thinking launched for Plus and Pro users. That gives power users a temporary escape hatch if they prefer the older reasoning model for certain workflows, prompts, or output style. ### Why does this matter beyond model nerds? (help.openai.com) Because the product is moving from explicit model choice to managed routing. Basically, ChatGPT is becoming more like an automatic transmission — easier for most people, but less transparent for anyone who cares which engine is actually running. That can be great for convenience. But it also means “use Instant” no longer tells you one exact thing about the system underneath. ### Bottom line? OpenAI is collapsing more of ChatGPT into one default experience. That makes the app simpler on the surface. But under the hood, model choice is getting more dynamic, more hidden, and more temporary. (help.openai.com)