OpenAI releases GPT-6.1 Terra, a fast model for computer use

On October 26, 2026, OpenAI released GPT-6.1 Terra in its API. OpenAI describes it as a fast, efficient model for visual reasoning, precise computer use, and clear, natural writing. It takes text and image input and produces text output.

In the API, GPT-6.1 Terra costs $1.00 per million input tokens and $5.00 per million output tokens. Its context window is 1,050,000 tokens, with up to 128,000 max output tokens.

OpenAI suggests four kinds of work for it: matching product photos to catalog entries, interpreting technical drawings, acting on application screens, and drafting and revising text.

If you build with the OpenAI API, a few pricing details change what you actually pay. Cached input costs $0.05 per million tokens, 5% of the uncached input rate, and Batch and Flex prices are 50% lower than Standard. The large context window has a catch: prompts with more than 272K input tokens are priced at 2x the input and cache rates and 1.5x output, for the full request.

If you want it to act on application screens, start with the Responses API: computer use is among the tools Terra supports there, and tool calling requires the Responses API.

Our take:

for screen automation on a budget, test Terra before you reach for a bigger model.