GPT-6.1 Terra Guide: Pricing, API, Codex and Sol Compared

Last updated: October 29, 2026

GPT-6.1 Terra is OpenAI’s fast, lower-cost model for visual reasoning, computer use and writing, released October 26, 2026. It is available in ChatGPT Work and Codex on Plus, Pro, Business, Enterprise and Edu plans, and in the API as gpt-6.1-terra at $1 per million input tokens and $5 per million output tokens.

OpenAI describes Terra as “a significant step up from GPT‑5.6 Terra” and says its standard API prices are 50% lower than those of GPT-5.6 Terra and GPT-6.1 Sol (OpenAI announcement). For background on the flagship model, see our explainer on what GPT-6 Astra is.

Contents

What’s new in GPT-6.1 Terra
How to use GPT-6.1 Terra in ChatGPT Work and Codex
How to call GPT-6.1 Terra in the API
GPT-6.1 Terra pricing
GPT-6.1 Terra vs GPT-6.1 Sol and GPT-6 Astra
Workflows and prompt patterns
Limitations to know
Frequently asked questions

What’s new in GPT-6.1 Terra

The table compares the two Terra generations using OpenAI’s GPT-6.1 Terra model page, the GPT-5.6 Terra model page and the API pricing page.

	GPT-6.1 Terra	GPT-5.6 Terra
Model ID	`gpt-6.1-terra`	`gpt-5.6-terra`
Context window	1,050,000 tokens	1,050,000 tokens
Max input / output	922,000 / 128,000 tokens	922,000 / 128,000 tokens
Knowledge cutoff	June 1, 2026	February 16, 2026
Reasoning effort	none, low, medium (default), high, xhigh, max	none, low, medium (default), high, xhigh, max
Standard price per 1M tokens (input / cached / output)	$1 / $0.05 / $5	$2 / $0.20 / $12

The larger changes are in results. In OpenAI’s launch post, Terra solves 50.3% of Terminal-Bench 4.0 tasks at maximum effort, against 21.5% for GPT-5.6 Terra. On OpenAI’s internal set of error-inducing prompts, the error rate at low effort fell from 53.6% to 16.7%. OpenAI also says Terra completes computer-use tasks more than three times faster than GPT-6 Astra and GPT-6.1 Sol. These are vendor figures, not independent tests.

How to use GPT-6.1 Terra in ChatGPT Work and Codex

Terra is not offered in the Chat tab. According to OpenAI’s ChatGPT and Codex models page, you pick it in Work or Codex:

In the ChatGPT desktop app or ChatGPT Work on the web, open Advanced in the model menu and choose GPT-6.1 Terra. Work also supports Terra on mobile.
In Codex CLI, update to version 0.159.0 or later, then start a session with codex -m gpt-6.1-terra.
Switch inside a session with /model, which also lets you change reasoning effort.
Run it non-interactively with codex exec -m gpt-6.1-terra "Review the current changes".
Set up browser use or computer use before you hand Terra screen-based tasks, because each has its own setup and access requirements.

OpenAI notes that some Plus and Pro users may see GPT-6.1 Terra at Medium instead of GPT-6.1 Sol at Light in the default Power presets in Work. Overall Work and Codex plan limits are unchanged.

How to call GPT-6.1 Terra in the API

Terra supports the Responses, Chat Completions and Batch endpoints, but tool calling requires the Responses API (model page). A minimal request looks like this:

curl https://api.openai.com/v1/responses \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-6.1-terra", "reasoning": {"effort": "medium"}, "input": "Summarize the attached invoice."}'

Follow these steps from OpenAI’s GPT-6 guide when you migrate:

Set model to gpt-6.1-terra in a Responses API request.
Start at medium effort and raise it only if the quality gain is worth the added time and tokens. Use reasoning.effort in Responses or reasoning_effort in Chat Completions.
Move tool calls to Responses. Terra supports web search, file search, code interpreter, hosted shell, apply patch, skills, computer use, MCP and tool search there.
Remove temperature, top_p and top_logprobs when reasoning effort is not none.
Compare against GPT-5.6 Terra on representative images and workflows at the same effort before switching production traffic.

GPT-6.1 Terra pricing

API prices come from OpenAI’s pricing page. ChatGPT credit rates and usage estimates come from the ChatGPT Work and Codex pricing page.

Option	Price or allowance
API Standard, up to 272K input tokens	$1 input, $0.05 cached input, $5 output per 1M tokens
API Standard, over 272K input tokens	$2 input, $0.10 cached input, $7.50 output per 1M tokens
Cache writes	$1.25 per 1M tokens (30-minute retention) or $2.50 (12-hour)
Batch and Flex	50% below Standard: $0.50 input, $2.50 output
Fast mode	2x Standard
ChatGPT credits	25 credits input, 1.25 cached, 125 output per 1M tokens
Plus local messages per five hours	About 35–285 (Sol: 15–160; Astra: 5–45)

The message ranges are OpenAI’s estimates for Plus and Standard Business, not fixed limits. Pro plans currently have no five-hour limit.

GPT-6.1 Terra vs GPT-6.1 Sol and GPT-6 Astra

	GPT-6.1 Terra	GPT-6.1 Sol	GPT-6 Astra
Standard price per 1M tokens (input / output)	$1 / $5	$2 / $10	$10 / $50
Lowest reasoning effort	none	low	low
Knowledge cutoff	June 1, 2026	April 30, 2026	—
OpenAI positioning	Fast and efficient: writing, visual work, computer use	Near-Astra performance at lower cost	Highest intelligence

Prices are from the pricing page and the GPT-6.1 Sol model page. A dash means this guide did not confirm the value. OpenAI says Terra at medium effort performs on par with GPT-6 Astra on Agents’ Last Exam at one-tenth its price per token, but that Sol and Astra remain the better choice for the most complex coding. Against Anthropic, OpenAI reports Terra at max effort is one-fourth the price per token of Claude Opus 5.5, which our Claude Code guide covers. For the history of the GPT-6.1 line, see our report on the planned GPT-6.1 Astra release.

Workflows and prompt patterns

OpenAI’s model selection guide recommends Terra for concise writing from supplied material, computer-use tasks with a defined outcome and scoped code fixes.

Read screenshots and charts. OpenAI’s sample prompt: “Review these dashboard screenshots and summarize the main changes. Refer to the chart labels and values, and flag anything you can’t read confidently.”
Automate form entry. Ask Terra to fill out a form from a source document, then reopen the record to confirm the saved values and check that unrelated data was not changed.
Match images to records. OpenAI’s example is matching supplier lookbook photos to catalog entries and updating listings.
A/B test against Sol on the same inputs and effort, and keep the lighter model if it meets your quality bar.

Limitations to know

Terra accepts text and images only. Audio and video input are not supported, and it outputs text only.
Fine-tuning and predicted outputs are not supported, and Realtime, Assistants and Embeddings endpoints are not available.
In ChatGPT it is limited to Work and Codex, and it does not support UAE inference residency.
For the most complex coding tasks, OpenAI recommends GPT-6.1 Sol or GPT-6 Astra.
If ChatGPT or Codex stops responding, check our AI service status guide first.

Frequently asked questions

Is GPT-6.1 Terra available in ChatGPT for free?

No. GPT-6.1 Terra is available in ChatGPT Work and Codex on Plus, Pro, Business, Enterprise and Edu plans. It is not available in the Chat tab, and Enterprise access depends on workspace availability and administrator controls.

How much does GPT-6.1 Terra cost in the API?

Standard pricing is $1 per million input tokens, $0.05 per million cached input tokens and $5 per million output tokens. Prompts over 272K input tokens cost 2x for input and 1.5x for output. Batch and Flex are 50% cheaper, and Fast mode costs 2x Standard.

What is the difference between GPT-6.1 Terra and GPT-6.1 Sol?

Terra costs half as much as Sol in the API ($1 and $5 versus $2 and $10 per million tokens) and supports the none reasoning effort, which Sol does not. OpenAI says Sol and GPT-6 Astra remain the better choice for the most complex coding tasks.

What is the GPT-6.1 Terra context window?

OpenAI’s model page lists a 1,050,000-token context window, a maximum of 922,000 input tokens and 128,000 output tokens. The knowledge cutoff is June 1, 2026. Terra accepts text and image input and produces text.

How do I use GPT-6.1 Terra in Codex CLI?

Update to Codex CLI 0.159.0 or later, then run codex -m gpt-6.1-terra. In an open session, use /model to switch to Terra or change its reasoning effort.

Sources

Image: OpenAI logo with magnifying glass, by Jernej Furman, CC BY 2.0, via Wikimedia Commons. Resized.

GPT-6.1 Terra: Power-User Guide to Pricing, API, Codex and Limits