What is GPT-6.1 Terra?
GPT-6.1 Terra is an OpenAI GPT-6.1 model for visual reasoning, computer use and natural writing at a lower token price than GPT-6.1 Sol. Use it to read invoices or product photos and act on application screens, or to draft and revise text. OpenAI still recommends GPT-6.1 Sol or GPT-6 Astra for the most complex coding.
Launch
OpenAI released GPT-6.1 Terra on October 26, 2026; the launch details and OpenAI's benchmark claims are in our news article, OpenAI releases GPT-6.1 Terra for computer use at half the Sol price.
What can you do with it?
OpenAI's documentation lists matching product photos to catalog entries, interpreting technical drawings, acting on application screens and drafting or revising text. A typical high-volume task from the announcement is reading incoming invoices and entering their details into accounting software through computer use.
For writing, Terra is positioned for proposals and everyday documents. For hard coding problems, OpenAI itself says GPT-6.1 Sol and GPT-6 Astra remain the better choice, so Terra fits routine edits rather than complex engineering work.
Try it on a small task
Suggested workflow: take 20 invoices or receipts you have already entered by hand. Send each image to gpt-6.1-terra at the default medium effort with a fixed list of fields to return as JSON, such as supplier, date, total and tax.
Compare every field with your own entries and count errors, then rerun the failures at high effort. Check the token usage the API reports and compare cost per correct invoice with GPT-6.1 Sol on the same files. If you are migrating from GPT-5.6 Terra, run the same files on both models at the same effort.
Pricing and access
In the API, the model ID is gpt-6.1-terra. Standard pricing is $1 per million input tokens, $0.05 for cached input and $5 per million output tokens. Cache writes cost $1.25 per million tokens for 30-minute retention and $2.50 for 12-hour retention. Once a prompt passes 272K input tokens, the entire request is billed at a long-context rate: input and cache prices double and output rises by half. Asynchronous Batch and Flex processing halves the standard rates, Fast mode doubles them, and regional processing, where offered, adds 10%.
In ChatGPT, OpenAI says Terra is available to Plus, Pro, Enterprise and Edu users in ChatGPT Work and Codex. OpenAI's help center says GPT-6.1 Terra is a Work and Codex model, not available in regular ChatGPT conversations. Tool calling in the API requires the Responses API, and fine-tuning is not supported.
What stands out
- 1.05M-token context window and 128K max output tokens
- Reasoning effort from none to max, medium by default
- Text and image input; no audio or video
- Responses API tools including computer use and hosted shell
Before you choose
- OpenAI recommends Sol or Astra for the most complex coding.
- Not available in regular ChatGPT chats, only in Work and Codex.
- No audio or video input.
Go to the source
Sources & further reading
Based on the official documentation below. Features and access can change; check the provider before making a decision.
Was this page useful?
Help us make our coverage more useful. This is feedback on the page, not a rating of the tool.
Go a little further