GPT-6.1 Terra is an OpenAI API model built to be fast and efficient while it understands images, operates app screens and writes clear text.
In plain words
OpenAI added Terra to its API in October 2026, beside GPT-6 Astra and GPT-6.1 Sol. It reads text and images and writes text, and it can hold about a million tokens of context. You choose how hard it thinks, from no extra thinking up to a maximum setting, and it can call tools such as web search, file search and computer use. It costs $1 per million input tokens and $5 per million output tokens.
A quick example
Imagine an online shop with thousands of product photos. A developer sends each photo to Terra, asks which catalogue entry it matches, and then checks the matches it returns before updating the shop.
Where you'll meet it
- Apps that read images, such as product photos or technical drawings.
- Agents that act on application screens.
- Drafting and revising text.
Related
Sources
- DocsGPT-6.1 Terra model page, OpenAI API docs · read 26 Oct 2026
- DocsModels, OpenAI API docs · read 26 Oct 2026
- DocsChangelog, OpenAI API docs · read 26 Oct 2026
Checked by WhatIsAI against the sources above · 26 Oct 2026. Spotted an error? Report it.