OpenAI has introduced GPT-6.1 Terra, a new model offering improved visual reasoning, computer use, and cost efficiency. It delivers clearer, more natural writing and excels at everyday work tasks like drafting proposals, interpreting technical drawings, and editing videos.
GPT-6.1 Terra’s standard API prices are 50% lower than GPT‑5.6 Terra and GPT-6.1 Sol, making it practical for developers automating high-volume work such as reading invoices and entering data into accounting software. In ChatGPT Work and Codex, the same usage allowance covers more tasks than previous versions.
On Agents’ Last Exam, a benchmark of professional tasks, GPT-6.1 Terra scores 15 percentage points higher than GPT‑5.6 Terra at about one-third the cost. At max effort, it is over three times as token efficient as Claude Opus 5.5 and one-fourth its price per token.
This year, OpenAI focused on reducing hallucinations and factual errors. The error rate at low reasoning effort dropped from 53.6% with GPT‑5.6 Terra to 16.7% with GPT-6.1 Terra. On Terminal-Bench 4.0, it solved 50.3% of coding tasks at max reasoning, more than double GPT‑5.6 Terra’s published score of 21.5%.
Source: OpenAI Blog · Summarized by HeadlinesBriefing