OpenAI Launches GPT-6.1 Terra, Its Half-Good AI Model at Half the Price

In brief OpenAI released GPT-6.1 Terra on October 26, with standard API prices 50% lower than GPT-5.6 Terra. It is available to Plus, Pro, Enterprise and Edu users in ChatGPT Work and Codex, and to developers as gpt-6.1-terra. In OpenAI's internal test, Terra's rate of factual errors at low reasoning effort fell from 53.6% to 16.7%, and at maximum effort it solved 50.3% of Terminal-Bench 4.0 tasks versus 21.5% for GPT-5.6 Terra. OpenAI released GPT-6.1 Terra on Monday, an upgrade to the middle model in its lineup that it says costs developers 50% less than GPT-5.6 Terra and handles computer tasks faster and more accurately.Terra is the everyday model, built for routine office work like reading what is on a screen, clicking through apps, and drafting replies. That skill is called computer use, meaning an AI that sees your screen and clicks and types the way you would. The 50% cut applies to standard API prices, the rates developers pay to plug the model into their own apps.Myriad: Which company will have the best AI model? Click to make your prediction.“It’s a significant step up over GPT-5.6 Terra, with 50% lower standard API prices,” OpenAI said in its announcement, adding that “GPT-6.1 Terra is rolling out now and is available to all Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex.”ChatGPT Work is OpenAI’s agent mode inside ChatGPT, which can work across apps and browse and complete tasks on websites for you, and Codex is OpenAI’s coding tool. Free users get to use the lower-tier GPT-5.6 Luna.On an internal OpenAI test built from prompts designed to trick models into making mistakes, the rate of factual errors at low reasoning effort, a setting that controls how long the model thinks, fell from 53.6% for GPT-5.6 Terra to 16.7%. At maximum effort it fell 18 points to 4.3%.On Terminal-Bench 4.0, which tests coding agents on tasks run through a computer’s command-line terminal and scores the share of tasks completed successfully, Terra at maximum effort solved 50.3% of tasks versus 21.5% for GPT-5.6 Terra. OpenAI also says Terra finishes tasks more than three times faster on average than GPT-6 Astra and GPT-6.1 Sol. All of the figures are OpenAI’s own, and the error test is an internal one.OpenAI introduced Terra on June 26 as part of the GPT-5.6 family in a limited preview requested by the Donald Trump administration, then opened it to the public on July 9 as a model competitive with GPT-5.5 at about half the cost.In September the company launched GPT-6 Sol and Luna minutes after Anthropic released Claude Opus 5.5, with no new Terra.For in-depth research or complex coding, OpenAI still points users to GPT-6.1 Sol or GPT-6 Astra.Daily Debrief NewsletterStart every day with the top news stories right now, plus original features, a podcast, videos and more. Join 72,000+ readersCreate an account to save your articles.