OpenAI has been achieving noteworthy advancements with its latest updates, such as GPT-5.2 and ChatGPT Images. This week, the organization rolled out an upgrade to its ChatGPT Android app, a modification that power users have been eagerly awaiting. The official ChatGPT application now includes a true Thinking mode, which was previously only available through a web browser.
According to Bleeping Computer, the initial Thinking toggle rerouted responses through the Standard thinking time, which utilized less computing resources and had a quicker turnaround. The updated toggle gives users the option to select among Auto, Instant, or Thinking modes, which dictates how long the AI will deliberate on a request prior to delivering a response.
This enhancement corresponds with the capability that desktop users have when choosing “Extended Thinking,” providing the AI with additional time to process intricate inquiries for more precise answers.
The rollout is presently in progress, so if you access ChatGPT on an Android device, make sure you’ve upgraded to the most current version. Following the update, the new Thinking mode will be accessible in your chat settings, permitting you to opt between Standard and Extended thinking times for improved responses on mobile. As of now, the new thinking toggle is not available in the ChatGPT iOS application.
This function is limited to ChatGPT Plus subscribers, leaving free users with the previous Thinking mode, which is primarily the Standard mode.
According to Bleeping Computer, this feature was launched together with a new formatting block feature. Now, when you instruct ChatGPT to draft an email, it does so within a formatting block, allowing you to request modifications or revise the document without having to regenerate the entire email. This feature started rolling out over the Christmas period and is accessible to all users.
It’s a minor yet delightful conclusion to the year for ChatGPT users. OpenAI reportedly spent the latter part of 2025 in a “code red” phase after losing a considerable number of users to Google’s Gemini 3 and Nano Banana, and has been striving to reclaim those users through GPT-5.2 and ChatGPT Images. This initiative has proven to be largely effective, as GPT-5.2 competes closely with Gemini 3 on numerous benchmark tests while outperforming Grok 4 in many aspects.
Disclosure: Ziff Davis, the parent company of Mashable, initiated a lawsuit against OpenAI in April 2025, claiming that it violated Ziff Davis copyrights in the training and functioning of its AI systems.