OpenAI has launched GPT-5.1, a major ChatGPT update featuring two new models—GPT-5.1 Instant and GPT-5.1 Thinking—designed to deliver smarter, more personalised, and engaging conversations.
CEO Sam Altman stated the models are “not only smarter but also enjoyable to talk to,” addressing previous feedback about robotic responses and excessive jargon.
Key Takeaways
- Two new models: GPT-5.1 Instant for speed and wit, GPT-5.1 Thinking for complex reasoning.
- Rollout starts November 13, 2025: Paid users get priority access.
- Enhanced personalisation: Eight tone presets and experimental fine-tuning sliders.
GPT-5.1 Instant: Speed Meets Personality
GPT-5.1 Instant is the new flagship model for everyday use. It’s described as warmer and more conversational, adding a touch of playfulness while maintaining clarity.
Ideal for brainstorming and casual chats, its key improvements include:
- Enhanced responsiveness for smoother conversational flow.
- Smart decision-making that balances speed and accuracy.
- User-friendly design that makes AI feel more like a companion.
These upgrades could streamline workflows for businesses and creators using ChatGPT for content or customer support.
GPT-5.1 Thinking: Adaptive Reasoning for Complex Tasks
This model is tailored for in-depth analysis, dynamically allocating “thinking time” based on query complexity.
Notable upgrades include:
- Clearer, jargon-free communication perfect for technical explanations.
- Greater precision in data interpretation and strategic planning, reducing errors.
It will be particularly useful for professionals in tech, education, and research.
Availability and Rollout Schedule
The GPT-5.1 update begins rolling out from November 13, 2025. Deployment is phased:
- First access: Pro, Plus, Go, and Business users.
- Enterprise & Education: Seven-day early access via toggle.
- Free users: Access in the coming days after paid tiers.
OpenAI also teased an upcoming upgrade from GPT-5 Pro to GPT-5.1 Pro.
Advanced Customisation Options
GPT-5.1 introduces robust tone and style controls with eight presets: Default, Friendly, Efficient, Professional, Candid, Quirky, Cynical, and Nerdy.
An experimental fine-tuning feature for select users offers sliders for conciseness, warmth, scannability, and emoji frequency, allowing users to craft a bespoke AI voice.
