Half the price, and rated Critical for cyber

OpenAI has released GPT-6.1 Terra, a fast and efficient model it describes as combining detailed visual reasoning and precise computer use with clearer writing. It is aimed at the communication and computer-based work that fills a working day: drafting proposals, interpreting technical drawings, filling in forms, editing video. Its standard API prices are 50 percent lower than those of GPT-5.6 Terra and GPT-6.1 Sol.

The company's own measurements carry the case, and they are its own. On Agents' Last Exam, a benchmark of professional tasks, Terra at medium reasoning effort scored 15 percentage points higher than GPT-5.6 Terra at about one-third the cost, and OpenAI says it performs on par with GPT-6 Astra at one-tenth its price per token. On Terminal-Bench 4.0, a command-line coding test, it solved 50.3 percent of tasks at maximum effort against 21.5 percent published for GPT-5.6 Terra. OpenAI still recommends GPT-6.1 Sol and GPT-6 Astra for the most complex coding.

The figure with the most practical weight concerns errors. On OpenAI's internal benchmark of error-inducing prompts, the error rate at low reasoning effort fell from 53.6 percent with GPT-5.6 Terra to 16.7 percent, and at maximum effort it fell by 18 percentage points to 4.3 percent.

The safety line deserves equal attention. OpenAI states that GPT-6.1 Terra meets the Critical threshold in cybersecurity under its Preparedness Framework and keeps the protections developed for GPT-6 Astra. A model offered as the cheap, fast option is therefore also one its maker places at the top of its own scale for cyber capability, which is the combination worth carrying into any decision about where to deploy it. Every figure in the announcement is OpenAI's, measured on benchmarks it chose to report.