OpenAI released GPT-6.1 Terra, a faster and cheaper model aimed at everyday computer work — reading invoices, filling forms, handling technical drawings and light editing workflows.
OpenAI says error rates on an internal computer-use benchmark fell sharply at low reasoning effort, while API pricing is reported at about 50% below prior Terra-class offerings. For the hardest coding workloads, the company still points teams to higher-end Sol and Astra variants.
The launch continues the industry trend toward tiered model stacks: premium frontier models for hard reasoning, and mid-tier “workhorse” models for agentic office automation.
Questions? Chat with our team — threaded comments are coming soon.