New Delhi: There has been a big explosion in the world of Artificial Intelligence. ChatGPT maker OpenAI has launched its new and most advanced AI model GPT-6.1 Terra. The company says that this is not just a question-answering chatbot, but your personal digital agent. OpenAI CEO Sam Altman himself gave this information on social media platform X. they wrote, "6.1 Terra is here! It is capable of many tasks, but we are most excited about this new era of computer use."
What will the new Terra model do?
Older AI models were limited to just typing text, but GPT-6.1 Terra can see and understand your screen like humans can. It can itself click with the mouse, type with the keyboard and complete the work by moving from one app to another. If you understand in simple language, it will do your everyday boring tasks automatically. Like entering bill details into accounting software, filling lengthy online forms, creating a business proposal or converting an architectural drawing directly into a 3D model. This will open the way for large-scale automation for companies.
GPT-6.1 Terra: delivering fast and capable computer use at a lower price. GPT-6.1 Terra understands what's on screen, takes action across apps, writes naturally, and is a more capable model for your whole workday with the efficiency to take on more of it. https://t.co/8JTN1yonmV
— OpenAI (@OpenAI) October 26, 2026
Three times faster and 50% cheaper
The company claims that this new model is a big upgrade from the previous version GPT-5.6 Terra. It works on average three times faster than the company's other high-end models GPT-6.1 Sol and GPT-6 Astra. Not only the speed, there has also been a big reduction in the price. Its API price for developers has been reduced by 50%. Now the charge for input token will be only $1 per million and the charge for output token will be $5 per million. This will prove to be very economical for big businesses.
Made fewer mistakes, became expert in coding
OpenAI has made this model more intelligent and less error prone. Its error rate in internal tests has come down from 53.6% to just 16.7%. Whereas in full reasoning mode this rate reduces to only 4.3%. It has done wonders in coding also. In the difficult terminal benchmark test, it has scored twice as much as the old model. For this reason, it is being launched immediately for programmers on GitHub Copilot. Currently this model is live on ChatGPT Plus, Pro, Enterprise, Edu, ChatGPT Work, Codex, OpenAI API and Amazon Bedrock.