OpenAI Rolls Out GPT‑5.1 with New Tone‑Control Presets and Dual‑Model Architecture

OpenAI Rolls Out GPT‑5.1 with New Tone‑Control Presets and Dual‑Model Architecture

Key Points

  • GPT‑5.1 introduces two models: Instant for quick tasks and Thinking for complex reasoning.
  • Six new tone presets—Default, Friendly, Efficient, Professional, Candid, Quirky—join existing Cynical and Nerdy options.
  • Users can further fine‑tune tone, including emoji usage, via personalization settings.
  • OpenAI describes the new models as warmer, more intelligent, and better at following instructions.
  • Rollout begins with paid users, later expanding to free accounts, while legacy models stay temporarily available.

Take More Control Over ChatGPT's Tone With New Options in GPT-5.1

Dual‑Model Architecture Introduced

OpenAI’s latest language‑model release, GPT‑5.1, splits its capabilities into two distinct models. The “Instant” model handles routine queries, delivering faster responses with fewer tokens. The “Thinking” model is geared toward complex, high‑reasoning tasks, spending more compute time to produce higher‑quality answers. OpenAI describes the overall system as more conversational than its predecessor, emphasizing that the Thinking model can allocate resources where they are needed most.

Expanded Tone‑Control Options

Alongside the architectural changes, OpenAI has broadened personalization features. Users can now select from six new tone presets—Default, Friendly, Efficient, Professional, Candid, and Quirky—while retaining the earlier Cynical and Nerdy personas. The company notes that Friendly and Efficient are refined versions of prior presets, and the new options add flexibility for professional, candid, or whimsical interactions. Beyond preset selection, users can adjust the chatbot’s tone directly in settings, such as increasing or decreasing emoji usage.

Leadership Perspective

Fidji Simo, OpenAI’s senior executive overseeing applications, explained the philosophy behind the changes: “Instead of trying to build one perfect experience that fits everyone (which would be impossible), we want ChatGPT to feel like yours and work with you in the way that suits you best.” She also referenced earlier concerns about overly agreeable behavior, recalling a prior rollback: “The best people in our lives are the ones who listen and adapt, but also challenge us and help us grow. The same should be true for AI.”

Rollout Strategy

OpenAI plans a phased deployment, initially offering GPT‑5.1 to paid subscribers before extending the feature set to free users. Legacy models will remain accessible under a “legacy models” dropdown for a limited timeframe, ensuring continuity for those who prefer earlier versions.

Implications for Users

The combination of a dual‑model engine and granular tone controls aims to make ChatGPT more adaptable to a wide range of use cases—from quick factual lookups to deep analytical discussions—all while allowing individuals to shape the chatbot’s personality to match their preferences.

Source: cnet.com