San Francisco: OpenAI has unveiled its GPT-5.1 family of artificial intelligence models, marking the first major update since releasing its frontier models in August. The new models GPT-5.1 Instant, GPT-5.1 Thinking, and GPT-5.1 Auto offer improved tonality, enhanced intelligence, and better adherence to user instructions.
According to the San Francisco-based company, all models now interact in a warmer, more empathetic manner. Initially, the update will be available to paid subscribers, with plans to extend access to free users in the future.
In a detailed blog post, OpenAI explained that GPT-5.1 Instant and GPT-5.1 Thinking have begun rolling out to ChatGPT Go, Plus, Pro, and Business users.
GPT-5.1 Thinking now more effectively adjusts its thinking time based on the question, spending more time on complex problems, less on simple ones.
Responses are clearer with less jargon. pic.twitter.com/oCUrEPKkpE
— OpenAI (@OpenAI) November 12, 2025
Developers will also be able to access these models via the API later this week, with Instant listed as GPT-5.1-chat-latest and Thinking as GPT-5.1. GPT-5 itself will remain accessible for three more months under the legacy models dropdown for paid users, after which it will be retired.
Older models such as GPT-4o and GPT-4.1 will continue to be available. GPT-5.1 Instant, which will become the default model for free users after rollout, has been upgraded to feel ‘warmer by default and more conversational.’
OpenAI noted that during testing, users were surprised by the model’s playful personality, a response to criticism that GPT-5 felt dry and robotic compared to GPT-4. The model also excels in following instructions and now includes adaptive reasoning, enabling it to pause and think before responding when necessary.
GPT-5.1 Thinking has been enhanced for clarity and efficiency. It avoids jargon, communicates ideas more clearly, and can adjust its reasoning time depending on the complexity of questions. GPT-5.1 Auto acts as an advanced auto-router, seamlessly switching between Instant and Thinking models based on the user’s prompt.
Additionally, OpenAI revamped ChatGPT’s personalization tool. The Default mode has been refined, while the Listener and Robots presets have been renamed Friendly and Efficient, respectively.
Three new presets, Professional, Candid, and Quirky, have also been added. Professional is precise and polished, Candid is direct and encouraging, and Quirky is imaginative and playful. Existing Cynical and Nerdy presets remain unchanged.
ALSO READ | DATE 2025: Where tech, AI, and investment converge
