OpenAI has announced the release of GPT‑5.1, an update to its GPT‑5 series that delivers improved reasoning, conversational style, and customization features for users across its platform. The rollout includes two upgraded models:
- GPT‑5.1 Instant – ChatGPT’s most-used model, now warmer, more playful, and better at following user instructions.
- GPT‑5.1 Thinking – OpenAI’s advanced reasoning model, optimized to provide faster responses for simpler queries while spending more time on complex problems.
According to OpenAI, GPT‑5.1 models are designed to be not only more capable but more enjoyable to interact with, with early testing showing significant improvements in clarity, tone, and instruction adherence.
Adaptive Reasoning and Improved Performance
GPT‑5.1 Instant now uses adaptive reasoning to determine when to pause and think before responding, improving accuracy for challenging queries while maintaining fast response times. The model has demonstrated better performance on coding and math benchmarks, including AIME 2025 and Codeforces.
GPT‑5.1 Thinking adapts response time based on question complexity, making technical explanations more understandable and reducing jargon. The model’s default tone has also been adjusted to be warmer and more empathetic, particularly for complex tasks or professional use cases.
Expanded Personalization Options
OpenAI has also introduced new tools to customize ChatGPT’s tone and style. Preset styles include Default, Friendly, Efficient, Professional, Candid, and Quirky, with legacy options like Cynical and Nerdy remaining available. Users can now also finetune characteristics such as warmth, conciseness, scannability, and emoji use, with preferences applying across all active chats in real time.
This approach allows ChatGPT to adapt more naturally to individual user preferences, enhancing engagement and usability. The system can even offer proactive suggestions to adjust tone based on the flow of conversation.
Rollout and Availability
GPT‑5.1 is rolling out gradually to paid users (Pro, Plus, Go, Business), with free and logged-out users gaining access shortly after. Enterprise and Edu plans will have a seven-day early-access toggle, after which GPT‑5.1 will become the default model.
The API release for GPT‑5.1 Instant and Thinking is scheduled later this week, with Instant available as gpt-5.1-chat-latest and Thinking as GPT-5.1. Legacy GPT‑5 models will remain available under a sunset period of three months for paid subscribers.
OpenAI describes GPT‑5.1 as a step toward making ChatGPT smarter, more user-friendly, and highly customizable, reflecting the company’s continued focus on frontier AI innovation and user personalization.
