OpenAI Rolls Out GPT-5.5 Instant as ChatGPT Default

OpenAI has introduced a new default model for ChatGPT, replacing GPT-5.3 Instant with the newly launched GPT-5.5 Instant. The update focuses on delivering more accurate responses while keeping the fast performance users expect.

The company says GPT-5.5 Instant is designed to reduce hallucinations—especially in sensitive areas like law, medicine, and finance—without sacrificing speed. This marks a key improvement as AI tools continue to be used for more serious and real-world applications.

The broader GPT-5.5 model itself was released just last month, with OpenAI highlighting upgrades in coding, reasoning, and general knowledge tasks. Early benchmarks suggest a noticeable leap in performance.

The model scored 81.2 on the AIME 2025 math test, a significant jump from 65.4 achieved by its predecessor. It also showed stronger multimodal reasoning, scoring 76 on the MMMU-Pro benchmark compared to 69.2 previously.

One of the biggest changes comes in how the model handles context. GPT-5.5 Instant can now connect with tools to reference past chats, uploaded files, and even Gmail data to provide more personalized responses. This feature is currently rolling out to Plus and Pro users on the web, with mobile support expected soon. OpenAI also plans to expand access to Free users, Go Business subscribers, and enterprise customers in the coming weeks.

Transparency is another focus of this update. ChatGPT will now display memory sources across all models, allowing users to see where information in responses comes from. Users will also have the option to remove outdated data or correct inaccurate context. Importantly, OpenAI clarified that shared chats will not reveal these memory sources to others, helping maintain privacy.

For developers, GPT-5.5 will be accessible via API under the “chat-latest” option. Meanwhile, GPT-5.3 will remain available for paid users, but only for a limited period of three months before being phased out.

The transition comes after mixed reactions to earlier model changes. When OpenAI retired GPT-4o, many users pushed back, saying they had grown attached to its conversational style. Some even described it as a “best friend” or a reflection of themselves. Despite the backlash, the model was officially discontinued in February 2026.

With GPT-5.5 Instant now taking center stage, OpenAI appears focused on balancing performance, personalization, and trust—key factors as AI becomes an everyday tool for millions of users worldwide.

Also read : Suspected Backdoor Found in Daemon Tools Attack