Customize GPT-5's Thinking Speed with OpenAI's Latest Feature

Published: 19 Sep 2025

Author: Precedence Research

Share :

OpenAI has just launched a new feature that offers more control over how ChatGPT can quickly respond and lower its thinking time to enhance its responses without time delay. According to the recent changes, subscribers with plus, pro, and business options can choose the "thinking time” of the model directly from the message composer on the web.

Previous versions of ChatGPT, specifically GPT-5 with its thinking capability, used to take longer to respond, as the model was engineered to act this way before responding with accurate and authentic answers to queries asked by users. However, many users are expecting more flexibility at work due to time constraints to balance out both conviction with speed at work for efficiency for every task they are assigned to perform.

Hence, by recognizing users' needs, OpenAI has made a new updated version where subscribers can opt between a variety of pacing modes, such as standard, extended, and light. The standard mode offers new default settings for every plan while providing a perfect blend of intelligence and speed. Extended mode provides detailed information, which was formerly the default for Plus users, and takes more time to generate responses by using the deep thinking feature. On the other hand, the light mode of ChatGPT, exclusive for pro users, generates prompt replies for quick questions involving tasks to be resolved. Moreover, Heavy is another mode that was also available for users with options such as exclusive to pro. It was designed for deeper reasoning questions with a slow pace but highly comprehensive responses. All these modes will be available for future web chats until they are changed manually by the user.

Artificial Intelligence company wrote on its X account, formerly Twitter, “We’ve heard users' feedback that GPT-5 Thinking can sometimes take too much time. Now Plus, Pro, and Business users can set the pace to match the moment”

Meanwhile, OpenAI has also introduced Codex, a new version of its AI, engineered specifically for software development. It is designed to assist coders to write, debug, and review code more precisely so that they can work independently and along with humans too for complex queries for better results.