KR — 전자신문

OpenAI Releases Cost-Efficient 'GPT-6.1 Terra' with Stronger Vision and Computer Control

AIOctober 27, 20261 min read

OpenAI has released a new AI model called GPT-6.1 Terra. It can understand images and screens and operate software, and it costs half as much to use as its predecessor.

What happened

On May 26 (local time), OpenAI officially launched GPT-6.1 Terra. The model can understand pictures and screens and control real apps. It is aimed at companies that want to automate office work with AI. Tasks include writing proposals, reading technical drawings, filling in forms, and editing videos.

Key numbers

Standard API price is 50% lower than GPT-5.6 Terra and GPT-6.1 Sol.
Price is $1 per 1 million input tokens and $5 per 1 million output tokens.
On the Agent Last Exam test, it scored 15 percentage points higher than GPT-5.6 Terra at medium reasoning settings.
Cost per task is about one-third of the older model.
Compared with GPT-6 Astra, it performs about the same on the same test, at one-tenth the price per token, and finishes tasks more than 3 times faster.

Background

OpenAI sells AI models in tiers. Larger models are stronger but cost more to run. Terra is positioned as a value model, meaning it offers good performance for the money. OpenAI says that for very hard coding jobs, GPT-6.1 Sol or GPT-6 Astra is a better fit.

Why it matters

Lower prices make it cheaper for businesses to use AI for daily work. Fewer factual mistakes also make the model more reliable. OpenAI rates the model's cybersecurity abilities as "Critical," its highest risk level, and applies the same safeguards it uses for Astra.

What to watch

Whether companies adopt Terra for office automation at scale.
How independent tests compare with OpenAI's own results.
How the model handles real-world tasks outside OpenAI's tests.
How security safeguards work for a model rated at the highest cyber risk level.

Glossary

OpenAI: An AI company that makes ChatGPT and AI models for developers and businesses.
API: A way for software developers to connect their own programs to an AI model.

Anthropic is burning gas to feed the AI boom

In its S-1 filing with the US SEC, Anthropic acknowledges that its data centers increasingly rely on natural gas, which is primarily methane, leaving its buildout exposed to rising fuel costs and supply constraints. The company also says some planned data centers will use on-site natural gas turbines and that rising global temperatures could raise its operating costs. Anthropic further warns that regulators could impose conditions or charges on data centers over grid impacts.

AI전자신문Oct 27

OpenAI Claims to Solve Math, Computer Science, and Physics Challenges, with Nobel Laureates Verifying the Results

OpenAI announced on the 26th (local time) that its AI models produced new proofs for three long-standing problems in mathematics, computer science, and physics, spanning harmonic analysis, computational complexity theory, and quantum field theory. Independent verification and interpretation will involve prominent researchers, including a Nobel physics laureate and proposers of key conjectures. OpenAI said it released the AI-generated proofs and Lean formalizations for math and computer science, while the physics results are being summarized by researchers directly at the request of outside scientists.

AITechCrunchOct 27

Anthropic’s IPO filing reveals its strategy for keeping enterprise customers hooked

Anthropic's S-1 filing shows more than half of customers spending at least $100,000 annually now use both the Claude platform and a first-party app such as Claude Code or Cowork, up from about 30% at the end of last year. The company reported net revenue retention of about 900% as of the end of the third quarter and said its "integrated approach" across models, developer platform, and applications drives this growth. The filing also flags headwinds, including cloud partners increasingly seeing Anthropic as a competitor.

AIThe VergeOct 26

Kevin Roose is AGI-pilled. Why aren’t you?

In a Decoder interview, Nilay Patel talks with journalist Kevin Roose about his new book, The AGI Chronicles, which reports on the companies racing to build artificial general intelligence. They also discuss Roose's move from The New York Times to independent work, including launching the podcast Machine Gods with Casey Newton, and the challenges media companies face when pivoting to video.