OpenAI updates GPT‑5 to GPT‑5.1: fixes tone and introduces 8 modes

OpenAI has launched GPT-5.1, updating its flagship AI model with two new versions: GPT-5.1 Instant and GPT-5.1 Thinking. These models aim to follow instructions more accurately and provide more engaging conversations.

The update comes three months after the launch of GPT-5, which users felt was sometimes “cold” and struggled with tasks that older versions handled better.

GPT-5.1 Instant powers most ChatGPT interactions, offering quicker and more natural dialogue. GPT-5.1 Thinking focuses on extended reasoning to solve complex problems. Both models can judge the difficulty of questions in real time, responding faster to easy queries while taking more time on harder ones.

Users can choose from eight personality modes: Default, Professional, Friendly, Candid, Quirky, Efficient, Nerdy, and Cynical. There are also controls for emoji use and response tone, which take effect immediately across all chats.

First access is given to paid users on Pro, Plus, Go, and Business plans, while free accounts gain access shortly after. Enterprise and Education users get a one-week preview. Legacy GPT-5 models remain available for three months.

OpenAI says GPT-5.1 performs better in coding and math tests like Codeforces and AIME 2025, and GPT-5.1 Thinking explains complex ideas in simple terms.

Fidji Simo, CEO of Applications at OpenAI, said the update balances intelligence and approachability, creating AI that is “not only smart but also enjoyable to talk to.” The new models will be gradually rolled out over the next few days.

Read More: Kuwait Introduces One of the World’s Fastest Visas, Issued in Just Five Minutes