GPT 6.1 Terra — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- GPT-6.1 Terra can match product photos to catalog entries, interpret technical drawings, act on application screens, and draft or revise text.
- GPT-6.1 Terra supports reasoning effort levels of none, low, medium, high, xhigh and max, with medium as the default.
- Responses API is required for tool calling with GPT-6.1 Terra; the GPT-6 guide provides prompting and migration guidance.
- GPT-6.1 Terra prices cached input tokens at 5% of the uncached input-token rate.
- GPT-6.1 Terra sets standard cache writes at $1.25 per million tokens for 30-minute retention and $2.50 for 12-hour retention, before long-context adjustments.
- GPT-6.1 Terra applies 2x input and cache rates plus 1.5x output pricing to the full request when prompts exceed 272K input tokens.
- GPT-6.1 Terra prices Batch and Flex at 50% below Standard, adds a 10% regional-processing premium where available, and charges 2x Standard in Fast mode.
Why it matters: API developers can cut GPT-6.1 Terra costs by 50% with Batch or Flex, but regional processing adds 10% and Fast mode costs 2x Standard. Long prompts are costlier too: above 272K input tokens, the full request is charged at 2x input/cache rates and 1.5x output, making context length a direct budgeting constraint.
Ask SkimNews
