GPT-5.1 Release Notes: Enhanced Intelligence and Conversational Tone

OpenAI has released GPT-5.1, an upgrade to the GPT-5 series that improves both the intelligence and the conversational style of ChatGPT. The update introduces two primary models—GPT-5.1 Instant and GPT-5.1 Thinking—alongside new tools for users to customize the tone and personality of the AI's responses.

GPT-5.1 Instant: Adaptive Reasoning and Better Instruction Following

GPT-5.1 Instant is designed to be the most used model in the ChatGPT ecosystem, now featuring a warmer, more conversational default tone. Key technical and functional improvements include:

  • Adaptive Reasoning: For the first time, GPT-5.1 Instant can decide when to "think before responding" to challenging questions. This capability has led to significant performance gains in math and coding evaluations, specifically AIME 2025 and Codeforces.
  • Improved Instruction Following: The model more reliably answers the specific questions asked by the user.
  • Communication Style: The model is designed to be more playful while remaining clear and useful.

GPT-5.1 Thinking: Dynamic Efficiency and Clarity

GPT-5.1 Thinking is an advanced reasoning model optimized for efficiency and ease of understanding. Its primary updates focus on how it allocates computational resources and communicates results:

  • Dynamic Thinking Time: The model now adapts its thinking time more precisely to the complexity of the task. On a representative distribution of ChatGPT tasks, GPT-5.1 Thinking is roughly twice as fast on the fastest tasks and twice as slow on the slowest tasks compared to GPT-5 Thinking (with thinking time set to Standard).
  • Increased Clarity: Responses now contain less jargon and fewer undefined terms, making complex technical concepts more approachable.
  • Tone Shift: The default tone is now warmer and more empathetic.

ChatGPT Personalization and Tone Controls

OpenAI is introducing more intuitive controls to allow users to shape ChatGPT's personality across all models. These settings apply globally and take effect immediately, including in ongoing conversations.

Expanded Personality Presets

The existing presets have been refined and expanded. The current options include:

  • Default: Balanced style and tone.
  • Professional: Polished and precise.
  • Friendly: Warm and chatty (formerly "Listener").
  • Candid: Direct and encouraging.
  • Quirky: Playful and imaginative.
  • Efficient: Concise and plain (formerly "Robot").
  • Nerdy: Exploratory and enthusiastic.
  • Cynical: Available under personalization settings.

Granular Characteristic Tuning

As an experiment rolling out to a limited number of users, OpenAI is testing the ability to tune specific characteristics directly from personalization settings, such as:

  • Response conciseness
  • Warmth
  • Scannability
  • Emoji frequency

Additionally, ChatGPT may proactively offer to update these preferences when it detects a user requesting a specific tone or style during a conversation.

Deployment and Availability

Rollout Schedule

GPT-5.1 Instant and Thinking are rolling out starting November 12, 2025. The sequence is as follows:

  1. Paid Users: Pro, Plus, Go, and Business users receive access first.
  2. Free and Logged-out Users: Access follows the paid tier.
  3. Enterprise and Edu Plans: These users receive a seven-day early-access toggle (off by default) before GPT-5.1 becomes the sole default model.

API Access

Both models will be available in the API later in the week of the release. GPT-5.1 Instant will be listed as gpt-5.1-chat-latest and GPT-5.1 Thinking will be released as GPT-5.1, both featuring adaptive reasoning.

Legacy Model Support

GPT-5 (Instant and Thinking) will remain available via a legacy models dropdown for paid subscribers for three months to allow for comparison and adaptation. This sunset period does not affect other legacy models.

Sources