Oct. 26, 2026 — OpenAI introduced GPT-6.1 Terra as a model built less for abstract frontier bragging rights than for the repetitive digital chores that consume a workday: reading invoices, navigating forms, interpreting technical drawings and editing video. The company says Terra combines visual reasoning, natural-language writing and precise action across applications at API prices 50% below GPT-5.6 Terra and GPT-6.1 Sol.¹

Sam Altman framed the release as the beginning of a broader shift, writing: “6.1 Terra is here! It is good at a lot of things, but we are especially excited for this new era of computer use.”Sam Altman (sama)² Greg Brockman made the more personal claim that Terra’s computer use had crossed “a qualitative threshold” and felt faster than he could work unaided.Greg Brockman (gdb)³

The performance case — OpenAI says Terra scores 15 percentage points above GPT-5.6 Terra on Agents’ Last Exam at medium reasoning effort, for roughly one-third of the cost, and completes tasks more than three times faster than GPT-6 Astra and GPT-6.1 Sol. It also reports a sharp reduction in errors on its internal test of error-inducing prompts, from 53.6% to 16.7% at low reasoning effort.¹

Outside OpenAI’s announcement, reporting has echoed the company’s central pitch: a fast, efficient system aimed at communication and computer-based tasks that fill ordinary professional workflows, now available through ChatGPT Work, Codex and the API.⁴

The catch — Terra is not being marketed as a universal replacement for OpenAI’s flagship systems. Even as the company highlights its 50.3% result on Terminal-Bench 4.0—more than double GPT-5.6 Terra’s published score—it recommends GPT-6.1 Sol and GPT-6 Astra for the most complex coding tasks. Terra’s promise, then, is not that every hard problem has become cheap, but that far more routine work may be ready for automation.¹