OpenAI generally does a pretty good job of taking user feedback seriously. Whether it's bringing back a favorite legacy model, toning down ChatGPT-5's personality or taking into account how much users like ChatGPT Voice, the company has no problems pivoting to try to keep users happy. And that's generally a good idea when Gemini has overtaken you in the App Store.
Now the company is making a new update by request. ChatGPT users now have more control over how long the chatbot "thinks" before generating a response, in a move designed to address one of the most common pieces of feedback. If you know, you know: sometimes GPT-5 is just too slow. Here's what the latest update announced on X, means for Plus and Pro users.
ChatGPT-5: Smarter pacing options
Starting today, ChatGPT Plus, Pro, and Business subscribers can choose how much time GPT-5 Thinking spends reasoning before answering. The new controls are available directly in the ChatGPT web composer, making it easy to adjust the AI’s pace to match the task at hand. Here’s what’s available to users now:
- Standard: The new default for all users. It balances speed with intelligence, aiming to provide thoughtful answers without the wait.
- Extended: The old default for Plus users. It gives GPT-5 more time to reason through complex prompts, at the cost of slower replies.
- Light (Pro only): The snappiest option. Great for quick facts, short answers, or when you just don’t want to wait.
- Heavy (Pro only): The deepest reasoning mode. Best for challenging problems or multi-step analysis where extra thought pays off.
Once you select a mode, your preference will remain in place for future chats until you change it.
This change is a big deal because it takes into account both users reliance on GOT-5 Thinking and workflows. One of GPT-5 Thinking’s biggest strengths is its ability to break down complicated questions step by step. But that same process can cause frustrating delays, especially for users who just want fast, conversational responses.
By giving people control over how long the model spends “thinking,” OpenAI is acknowledging that no one setting works for every scenario.
In practice, this means you might keep GPT-5 on Light for quick brainstorming or casual chats, switch to Standard for everyday problem-solving, and only tap Heavy when you need the AI to dig deep into complex reasoning.
The takeaway
Although this marks a small update, it is an important one, underscoring how OpenAI approaches customization. Rather than setting a one-size-fits-all pace, the company is letting users decide what’s most useful in the moment. That flexibility goes a long way.
With customizable thinking times, ChatGPT now adapts more closely to how you work. Whether you value speed, depth or a balance of both, GPT-5 Thinking can now be set to match the moment.
Follow Tom's Guide on Google News and add us as a preferred source to get our up-to-date news, analysis, and reviews in your feeds. Make sure to click the Follow button!
