OpenAI has expanded its GPT-5 series with the launch of GPT-5.1, offering improved intelligence, clearer communication and easier customisation within ChatGPT. The update begins rolling out from today, starting with paid users, and is designed to make interactions both smarter and more natural.
The new release includes two updated models: GPT-5.1 Instant and GPT-5.1 Thinking. OpenAI describes the Instant model as its most widely used version of ChatGPT, now designed to feel warmer and more conversational. GPT-5.1 Thinking, which is built for more complex reasoning, has been refined to produce clearer explanations and deliver faster responses on simpler tasks.
OpenAI says the update reflects clear feedback from users who want AI that is not only capable but engaging. The company explains that people value intelligence and accuracy, but they also expect an AI assistant that communicates in an approachable and enjoyable way. GPT-5.1 aims to close this gap by improving both cognitive performance and tone.
New controls for conversation style
Alongside the updated models, OpenAI is introducing improved tools that allow users to shape ChatGPT’s tone more easily. The company notes that preferences vary widely, and even a single user may want different styles depending on the task at hand. To address this, ChatGPT now includes more intuitive settings that guide how it should sound, making it easier to match the style required for each conversation.
OpenAI says these controls are meant to give users more influence without needing to write highly specific prompts. The changes are part of a broader effort to make ChatGPT more responsive to personal and situational needs.
GPT-5.1 Instant becomes more conversational
The biggest update comes to GPT-5.1 Instant, which is built for fast, everyday queries. According to OpenAI, the model is now warmer by default and adopts a more conversational speaking style. The company says early testers found it surprisingly playful while still maintaining clarity and usefulness.
OpenAI believes these improvements make the AI feel more human-like in interactions, reducing friction in tasks that range from quick factual questions to lighter, more informal exchanges. While the company has not detailed the underlying technical changes, the update appears focused on adjusting the model’s communication behaviour rather than purely its reasoning capabilities.
The GPT-5.1 Thinking model, meanwhile, aims to stay persistent on multi-step or complex problems. OpenAI notes that it has become easier to understand, providing clearer reasoning and improved structure in its explanations.
