OpenAI is finally giving Android users the “keys to the engine.” As of Wednesday, December 31, 2025, the ChatGPT app has officially added a manual Thinking mode toggle, ending the era where the mobile app just “faked it” while the desktop got all the deep-reasoning power.
The thing is, before this update, the mobile “Thinking” state was mostly a UI trick. Or nothing. Let’s be real, it was routing your tough questions through the same standard compute as basic chats. Those too.
Also Read | Imran Khan and Bushra Bibi Sentenced to 17 Years in Jail
The “Thinking Depth” Update: Field Notes
If you’re a ChatGPT Plus subscriber, you can now explicitly tell the AI how hard to work on a problem.
-
Auto: The default. It guesses if you need a quick fact or a deep dive.
-
Instant: Pure speed. Best for “what’s the capital of France?” or quick translations.
-
Extended Thinking: This is the desktop-level “High Juice” mode. The AI will literally pause for 10–30 seconds to run through multi-step logic.
-
The “Go” Catch: If you’re on the cheaper ChatGPT Go plan (popular in India), you’re still locked out of this toggle for now.
Also Read | Imran Khan and Bushra Bibi Sentenced to 17 Years in Jail
The “Christmas Gift”: Formatting Blocks
And here’s the kicker: OpenAI quietly rolled out Formatting Blocks on Christmas Day.
-
The Problem: Before, if you asked for an email, it just appeared as a block of text in the chat bubble.
-
The Fix: Now, emails, blogs, and documents appear in a dedicated “rich text” block.
-
The Best Part: You can edit the text directly in the block using a mini-toolbar (bold, italics, lists) without having to ask the AI to “regenerate” the whole thing because you wanted one word changed.
Also Read | Imran Khan and Bushra Bibi Sentenced to 17 Years in Jail
OpenAI’s “Code Red” Year-End
It’s an ongoing situation where OpenAI spent most of 2025 looking over its shoulder. After losing a chunk of users to Gemini 3 and Google’s Nano Banana model earlier this year, the company went into a “Code Red” state.
These updates—along with the release of GPT-5.2 on December 11—are the closing moves in a desperate bid to reclaim the “Expert-Level” crown. GPT-5.2 is currently trading blows with Gemini 3, especially in spreadsheet formatting and financial modeling.
Basically, ChatGPT is trying to look less like a chatbot and more like a professional workstation as we head into 2026.
Also Read | Imran Khan and Bushra Bibi Sentenced to 17 Years in Jail
End…
