OpenAI rolls out GPT-5.1 to ChatGPT with sharper instruction following and adaptive reasoning
OpenAI has released GPT-5.1, the latest iteration of its ChatGPT models, featuring new reasoning capabilities and expanded personalization tools designed to improve both accuracy and usability.
The company says the update refines two key versions, GPT-5.1 Instant and GPT-5.1 Thinking, each optimized for different levels of task complexity.
According to OpenAI’s blog, GPT-5.1 introduces adaptive reasoning for the first time. The model now adjusts its thinking time based on question difficulty—spending more time on complex problems while responding faster to simpler ones. Internal benchmarks show significant performance improvements in math and coding evaluations, including AIME 2025 and Codeforces.
Nick Turley, Head of ChatGPT at OpenAI, said in a Linkedin post GPT-5.1 “brings improvements to how enjoyable ChatGPT feels to talk to, and how well it follows what you’re actually asking.” He added that GPT-5.1 Instant is “warmer, more reliable with instructions, and can use reasoning on tougher questions for the first time,” while GPT-5.1 Thinking “adapts its reasoning time to the complexity of the task and gives clearer, more approachable responses.”
Smarter instruction handling and personalization tools
User feedback pointed to inconsistency in tone and adherence to prompts. OpenAI says GPT-5.1 now follows instructions more closely, particularly on multi-step or contextual requests. The update also introduces expanded tone controls, letting users choose between Professional, Friendly, Candid, Efficient, and Quirky response styles.
These personalization features apply across all models and can be fine-tuned further through settings that control warmth, conciseness, or emoji use. ChatGPT will also proactively offer to adjust tone based on conversational cues.
Education and enterprise applications
For education and enterprise users, these changes could make interactions with ChatGPT more predictable and customizable. In classroom and training contexts, adaptive reasoning may improve clarity when explaining complex concepts, while tone controls allow educators to tailor responses for age or audience.
For businesses using ChatGPT through Enterprise or Edu plans, the update also promises more consistency across teams. Sidharth Sharma, GTM APAC at OpenAI, shared on Linkedin that GPT-5.1 is “more reliable with instructions” and aims to “respond in a way that feels right for you.”
Rollout and model access
OpenAI says the update begins rolling out this week to ChatGPT Plus, Pro, Business, and Enterprise users before reaching free and logged-out users. The models will also be made available via API—GPT-5.1 Instant as gpt-5.1-chat-latest and GPT-5.1 Thinking under GPT-5.1—later in the week.
Existing GPT-5 models will remain available for a three-month transition period before being retired. OpenAI notes that GPT-5.1 will become the new default model once rollout is complete.
Turley added: “All of this work continues to be a work in progress, and we’re looking forward to learning from feedback.”
The ETIH Innovation Awards 2026
The EdTech Innovation Hub Awards celebrate excellence in global education technology, with a particular focus on workforce development, AI integration, and innovative learning solutions across all stages of education.
Now open for entries, the ETIH Innovation Awards 2026 recognize the companies, platforms, and individuals driving transformation in the sector, from AI-driven assessment tools and personalized learning systems, to upskilling solutions and digital platforms that connect learners with real-world outcomes.
Submissions are open to organizations across the UK, the Americas, and internationally. Entries should highlight measurable impact, whether in K–12 classrooms, higher education institutions, or lifelong learning settings.
