What happened
On October 26, 2026, OpenAI introduced GPT-6.1 Terra, a faster model aimed at computer use, visual reading, and ordinary writing. It is available to Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex, and to developers as gpt-6.1-terra. Standard prices are $1 per million input tokens, $0.05 per million cached input tokens, and $5 per million output tokens. OpenAI says that is 50 percent lower than GPT-5.6 Terra and GPT-6.1 Sol, and that the same plan allowance covers more tasks than those models or GPT-6 Astra. On Terminal-Bench 4.0 at maximum reasoning, OpenAI reports 50.3 percent, more than double GPT-5.6 Terra's published 21.5 percent. On an internal set of error-inducing prompts, the error rate at low effort fell from 53.6 percent to 16.7 percent, and at maximum effort it fell 18 points to 4.3 percent. On Agents' Last Exam at medium effort, OpenAI says Terra is 15 points above GPT-5.6 Terra at about one-third the cost, on par with GPT-6 Astra at about one-tenth the price per token, and, at max effort, about nine times cheaper per task than Claude Opus 5.5. OpenAI also says it finishes those tasks more than three times faster than GPT-6 Astra and GPT-6.1 Sol.
Why it matters
A model that is merely cheaper is a discount. A model that is cheaper and posts a large move on the lab's own coding and error tests changes which jobs are worth automating at volume: invoices, forms, drawings, and desktop steps. OpenAI is also explicit about the ceiling. For the most complex coding, it still names GPT-6.1 Sol and GPT-6 Astra.
OpenAI says Terra meets the Critical threshold in cybersecurity under its Preparedness Framework and keeps the protections built for GPT-6 Astra. The announcement points to a system card for the prompt-injection tests.
On MMMU-Pro, OpenAI says Terra beats GPT-5.6 Terra at every reasoning setting and nearly matches GPT-6 Astra and GPT-6.1 Sol. The announcement does not print the raw MMMU-Pro scores.
On OSWorld 2.1 at medium effort, OpenAI says Terra passes GPT-5.6 Terra's best score at a fraction of the cost. The raw OSWorld score is not in the prose.