OpenAI releases GPT-6.1 Terra in ChatGPT, Codex and the API
OpenAI released GPT-6.1 Terra on Monday, October 26, making it available in ChatGPT Work, Codex and its developer API. In the API it costs half as much per input token as GPT-5.6 Terra, and OpenAI says it is built for visual reasoning, precise computer use and clear writing.
What GPT-6.1 Terra is for
OpenAI describes GPT-6.1 Terra as "a fast, efficient model for visual reasoning, precise computer use, and clear, natural writing." Its developer documentation gives examples such as matching product photos to catalog entries, interpreting technical drawings, acting on application screens and drafting or revising text.
It sits third in OpenAI's current lineup. GPT-6 Astra is what OpenAI calls its most intelligent model, GPT-6.1 Sol is a cheaper option for complex work, and OpenAI calls GPT-6 Luna its fastest and most cost-effective model. OpenAI files Terra under "fast and efficient."
Where you can use it
According to OpenAI's ChatGPT changelog, GPT-6.1 Terra is available in ChatGPT Work and Codex on the Plus, Pro, Business, Enterprise and Edu plans. Users pick it from the model list under Advanced in the ChatGPT desktop app or in ChatGPT Work on the web, and Work also supports it on mobile.
The model isn't available in regular ChatGPT chats. OpenAI also says Enterprise access depends on workspace availability and administrator settings.
API pricing and specs
For developers, the model ID is gpt-6.1-terra. It works with the Responses and Chat Completions APIs and the Batch API, though tool calling requires the Responses API. It accepts text and images and produces text only, with no audio or video support.
| Model | Input (per 1M tokens) | Output (per 1M tokens) |
|---|---|---|
| GPT-6.1 Terra | $1 | $5 |
| GPT-5.6 Terra | $2 | $12 |
| GPT-6.1 Sol | $2 | $10 |
Cached input costs $0.05 per million tokens, down from $0.20 for GPT-5.6 Terra. Prompts longer than 272,000 input tokens are billed at twice the input rate and 1.5 times the output rate, while Batch and Flex processing are 50 percent cheaper than standard pricing.
The model has a context window of 1,050,000 tokens, accepts up to 922,000 input tokens and can return up to 128,000 output tokens. Its knowledge cutoff is June 1, 2026. Developers can set its reasoning effort anywhere from none to max, with medium as the default.
What it means for you
If you pay for ChatGPT, GPT-6.1 Terra is a new option in ChatGPT Work and Codex for tasks that involve screenshots, images or controlling apps on your computer. Free accounts aren't on OpenAI's list, and the model won't appear in the normal chat view.
For developers, the lower price is the main story. Anyone running GPT-5.6 Terra can halve input costs and pay well under half as much for output. OpenAI recommends testing both models on the same images and workflows before switching, and moving any tool calls from Chat Completions to the Responses API.