Introducing GPT-6.1 Terra | OpenAI
OpenAI's new GPT-6.1 Terra is a faster, cheaper model aimed at high-volume work like document reading, form filling, and coding tasks, with API pricing half that of GPT-5.6 Terra. The headline claims are better visual reasoning, far fewer factual errors, and a big jump on coding benchmarks, though the company says its top models still win on the hardest tasks.
- Cost-performance: API pricing is 50% below GPT-5.6 Terra and GPT-6.1 Sol, and at medium effort it scores 15 points higher than GPT-5.6 Terra at about a third of the cost.
- Error rate: On OpenAI's internal error-inducing prompts, the low-effort error rate dropped from 53.6% to 16.7%, and at max effort it hit 4.3%.
- Coding: On Terminal-Bench 4.0, it solves 50.3% of tasks at max effort, more than double GPT-5.6 Terra's 21.5%, but OpenAI still recommends GPT-6.1 Sol or GPT-6 Astra for the most complex coding.
GPT-6.1 Terra is OpenAI's new fast, cheap model aimed at the routine computer work that fills most office days: drafting proposals, reading technical drawings, filling in forms, editing video. It succeeds GPT-5.6 Terra, and the pitch is cost and speed as much as raw capability. API pricing is $1 per million input tokens and $5 per million output, with cached input at $0.05, which OpenAI says is 50% below both GPT-5.6 Terra and GPT-6.1 Sol. It's available now to Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex, and to developers as `gpt-6.1-terra`. The company's own example for developers is automating high-volume work like reading incoming invoices and entering the details into accounting software.
The benchmark claims are where the numbers get interesting, though they come from OpenAI's own reporting. On Agents' Last Exam, a set of professional tasks such as turning architectural drawings into 3D models, Terra scores 15 points higher than GPT-5.6 Terra at medium effort for about a third of the cost, and roughly matches GPT-6 Astra at a tenth of its per-token price. It finishes tasks over three times faster than both Astra and Sol. On Terminal-Bench 4.0, a command-line coding test, Terra solves 50.3% of tasks at maximum effort, more than double GPT-5.6 Terra's 21.5%. OpenAI still recommends Sol or Astra for the hardest coding work. The factuality improvement is the most striking single figure: on the company's internal error-inducing prompts, the error rate at low effort dropped from 53.6% to 16.7%, and at maximum effort it fell to 4.3%.
On safety, the article says Terra meets the "Critical" threshold for cybersecurity under OpenAI's Preparedness Framework, which is the top tier of that framework's risk ratings, and that it keeps the protections built for Astra. OpenAI reports better scope adherence and more honest disclosure of limitations than GPT-5.6 Terra, plus strong resistance to prompt injection in its own tests. These are self-reported evaluations, and the full system card linked from the post is the place to check the details rather than the summary alone.