Who can use GPT-6.1 Terra in ChatGPT?

GPT-6.1 Terra launched on October 26, 2026 for ChatGPT Plus, Pro, Enterprise and Edu users in ChatGPT Work and in Codex, according to The Next Web. Developers can call GPT-6.1 Terra in the OpenAI API under the model id gpt-6.1-terra on the Chat Completions, Responses and Batch endpoints.

How does GPT-6.1 Terra compare to GPT-5.6 Terra?

OpenAI says GPT-6.1 Terra costs half as much per token as GPT-5.6 Terra and scored 15 percentage points higher on an image test. On OpenAI's set of error-inducing prompts, the error rate at low reasoning effort fell from 53.6% with GPT-5.6 Terra to 16.7% with GPT-6.1 Terra, and to 4.3% at maximum effort.

Should I use GPT-6.1 Terra for hard coding tasks?

OpenAI still recommends GPT-6.1 Sol or GPT-6 Astra for the hardest coding work. The Next Web reports that GPT-6.1 Terra solved about half of OpenAI's command-line coding tasks, more than double its predecessor, but its main focus is computer use, visual reasoning and writing rather than top-end coding.

Are there extra costs for long prompts on GPT-6.1 Terra?

Yes. OpenAI's model page says GPT-6.1 Terra prompts over 272K input tokens cost 2x the input and cache rates and 1.5x the output rate for the whole request. Batch and Flex processing are 50% cheaper than Standard, regional processing adds 10%, and fast mode costs 2x Standard.

OpenAI · 2026-10-26 · major

GPT-6.1 Terra — OpenAI's mid-tier computer-use model at half the price

GPT-6.1 Terra is OpenAI's new mid-tier model for computer use, visual reasoning and writing. It replaces GPT-5.6 Terra at $1/$5 per 1M tokens, half the old price, and is live in the API, ChatGPT Work and Codex.

OpenAI developer docs card for the GPT-6.1 Terra model

OpenAI's cheaper Terra tier moves to GPT-6.1 and is built to read screens and act across apps.

Quick facts

Maker	OpenAI
API model id	gpt-6.1-terra
Price	$1 input / $5 output per 1M tokens
Context window	1.05M tokens (128K output)
Knowledge cutoff	June 1, 2026
Availability	API, ChatGPT Work, Codex
Replaces	GPT-5.6 Terra

Pricing

Input	$1.00 / 1M tokens
Cached input	$0.05 / 1M tokens
Output	$5.00 / 1M tokens
Batch / Flex · vs Standard	50% off

source ↗

What is it?

GPT-6.1 Terra is the new mid-sized model in OpenAI's GPT-6 family, launched on October 26, 2026 as the successor to GPT-5.6 Terra. OpenAI's four tiers now run Luna, Terra, Sol and Astra, from cheapest to most capable. OpenAI describes Terra as a fast, efficient model for detailed visual reasoning, precise computer use and natural writing.

How does it work?

For computer use, the model reads what is on screen and acts by clicking, typing and moving between apps; on the Responses API it can call hosted tools such as computer use, code interpreter, hosted shell, MCP and tool search. It takes text and image input, returns text, and lets developers set reasoning effort from none up to max.

Why does it matter?

Office-style agent work such as filling forms or entering invoices can now run on a model that, per OpenAI's own tests, matched GPT-6 Astra on a visual test at about a tenth of the price. OpenAI still recommends Sol or Astra for the hardest coding work, and rates Terra at the Critical level for cybersecurity under its Preparedness Framework.

Who is it for?

developers building computer-use and office agents

Frequently asked questions

Who can use GPT-6.1 Terra in ChatGPT?: GPT-6.1 Terra launched on October 26, 2026 for ChatGPT Plus, Pro, Enterprise and Edu users in ChatGPT Work and in Codex, according to The Next Web. Developers can call GPT-6.1 Terra in the OpenAI API under the model id gpt-6.1-terra on the Chat Completions, Responses and Batch endpoints.
How does GPT-6.1 Terra compare to GPT-5.6 Terra?: OpenAI says GPT-6.1 Terra costs half as much per token as GPT-5.6 Terra and scored 15 percentage points higher on an image test. On OpenAI's set of error-inducing prompts, the error rate at low reasoning effort fell from 53.6% with GPT-5.6 Terra to 16.7% with GPT-6.1 Terra, and to 4.3% at maximum effort.
Should I use GPT-6.1 Terra for hard coding tasks?: OpenAI still recommends GPT-6.1 Sol or GPT-6 Astra for the hardest coding work. The Next Web reports that GPT-6.1 Terra solved about half of OpenAI's command-line coding tasks, more than double its predecessor, but its main focus is computer use, visual reasoning and writing rather than top-end coding.
Are there extra costs for long prompts on GPT-6.1 Terra?: Yes. OpenAI's model page says GPT-6.1 Terra prompts over 272K input tokens cost 2x the input and cache rates and 1.5x the output rate for the whole request. Batch and Flex processing are 50% cheaper than Standard, regional processing adds 10%, and fast mode costs 2x Standard.

Try it

model: gpt-6.1-terra (OpenAI API, Responses or Chat Completions)