Model: GPT-6.1 Terra
On 26 October 2026, OpenAI began rolling out GPT-6.1 Terra in ChatGPT Work and Codex and in the OpenAI API as gpt-6.1-terra — a capable, keenly priced model for visual reasoning, computer use and writing. Here’s what a cheaper, more capable model running across professional and developer surfaces means for how AI assistants read the web and the brands they choose to name.
Source: ChatGPT release notes — GPT-6.1 Terra in ChatGPT Work and Codex (26 October 2026)
- On 26 October 2026, OpenAI began rolling out GPT-6.1 Terra in ChatGPT Work and Codex and in the OpenAI API.
- Terra handles visual reasoning, computer use and writing, and lands at a low API price of 1 dollar per million input tokens and 5 dollars per million output.
- A cheaper, capable model means more of the agentic reading and answering layer can run on it — across apps you never see.
- In ChatGPT it lives in Work and Codex, not the Chat tab, so it shapes how professionals research vendors and tools.
- reconnAI tracks GPT-6.1 Terra alongside ChatGPT, Claude, Gemini, Perplexity, Copilot and Google AI Overview, and re-baselines when a major model ships.
What OpenAI actually shipped
On 26 October 2026, OpenAI began rolling out GPT-6.1 Terra in ChatGPT Work and Codex and, at the same time, in the OpenAI API as gpt-6.1-terra. OpenAI describes it as a model that “supports a range of tasks, including visual reasoning, computer use, and writing” — a broad, general-purpose model rather than a narrow specialist.
The rollout reaches Plus, Pro, Business, Enterprise and Edu plans, and you select GPT-6.1 Terra from the model list as it arrives. Usage limits are unchanged, Standard and Fast modes are supported where available, and some Plus and Pro users will see Terra at Medium in the default Power presets in Work. In the API, Terra is priced at 1 dollar per million input tokens and 5 dollars per million output tokens.
Terra, Sol and the GPT-6.1 line
Terra is the newest entry in the GPT-6.1 line, following GPT-6.1 Sol, the agentic sibling built for coding and computer use. Where Sol is tuned to act, Terra reads as a well-rounded workhorse — strong on visual reasoning and writing, priced to run at volume, and aimed squarely at everyday professional and developer tasks.
That continues a pattern we have tracked all autumn: vendors widening their families so different models answer different questions. We saw it with the GPT-6 Pro parallel-agent update and the GPT-6 Astra flagship, and across rivals with Claude Opus 5.5, its workhorse Sonnet 5.5 and Gemini 4 Argon. The model doing the work shapes the answer your customers see.
Why a cheaper API model matters
Terra’s low API price is the quiet headline. At 1 dollar per million input tokens and 5 dollars per million output, a capable model becomes cheap enough to sit behind a great many products — the assistants, chatbots, research tools and autonomous agents that increasingly answer questions on a brand’s behalf without ever carrying the ChatGPT name.
When the cost of a capable model falls, more of the answering layer runs on it, and it runs more often. That means more automated reading of your category, more multi-step retrieval, and more moments where a model decides whether to name you. Cheaper inference does not just save money; it multiplies how many answers get generated in the first place.
Why visual reasoning changes answers
Terra leads on visual reasoning and computer use, which matters because so much brand information lives in images: screenshots, pricing tables rendered as graphics, product shots, charts and PDFs. A model that reads what it sees, not just the raw text, can pull facts from pages that text-only systems skip — and can just as easily misread a cluttered or inconsistent one.
That raises the bar on clarity. The same discipline that helps a human skim your site — legible layouts, labelled figures, text that is not trapped inside an image — now helps a model read you correctly. It is the retrieval-first mindset we unpacked in how LLMs decide which brands to cite.
How agents read your site
When a capable model evaluates your category, it moves through a few steps in quick succession. Fetch. It pulls your pages, docs and profiles. Parse. It extracts who you serve, what you offer and on what terms. Corroborate. It cross-checks those claims against third-party sources. Decide. It names the providers it can stand behind.
Each step rewards structure. Clean headings, plain statements of fact and consistent details across your site and off-site sources make you easy to parse and safe to cite — the discipline at the heart of building a GEO strategy and Google’s own llms.txt guidance.
What the GPT-6.1 Terra launch changes
OpenAI began rolling out GPT-6.1 Terra on 26 October 2026.
Where GPT-6.1 Terra shows up
In ChatGPT, Terra lives in Work and Codex rather than the Chat tab, so the first people it answers for are professionals researching vendors, suppliers and tools inside their workspaces. That is a high-stakes surface: the shortlist a model builds in Work can feed a real buying decision.
But model availability rarely stays in one place. Terra is in the API from day one, and capable models tend to spread quickly into consumer tools — the way judgements fan out across Perplexity’s multi-model line-up and the broader shifts we covered in the ChatGPT Intelligent UI update. A model that starts in professional tooling often ends up answering ordinary questions elsewhere.
What this means for your content
The response is not to chase one model but to make your expertise unmistakable to any careful agent. Spell out who you help, the problems you solve, your credentials and the evidence behind your claims, in clean structured content a model can read without guessing — and make sure the facts in your images are also written as text.
Consistency does the heavy lifting: when your key facts line up everywhere a model looks, it can cite you without hedging. That is the readiness we set out in is your brand ready for AI search? and saw sharpen answers in how stronger models tighten recommendations.
Does GPT-6.1 Terra cite sources?
In browsing and agentic contexts, Terra can reference the pages it reads directly, so a brand with clear, current content can show up by name in the trail of an answer. When it is not browsing, it still leans on the brands and sources it has learned to associate with a topic — which is why consistent, well-structured evidence pays off whether or not a given answer shows its working.
Either way the influence is real, and a more capable, cheaper model makes the competition more about substance than tactics. You are competing to be the source a careful agent reaches for, across far more answers than before.
How to check your AI visibility
Because capable models spread quickly once they ship, it is worth taking a baseline now. Here is a short, practical way to see how today’s models handle the questions around your brand — today is 28 October 2026:
- List your real questions. Write down the questions your customers actually ask before choosing you.
- Ask the models directly. Run those questions in ChatGPT and record which brands, providers and sources get named.
- Note the framing. Watch how it describes you, who it mentions alongside you, and what it caveats.
- Check the sources. Where it browses, see which pages it leans on and whether your own content is among them.
- Compare across models and regions. Repeat in Gemini, Claude and Perplexity, and in your key markets, since answers differ.
- Track the trend. One run is a snapshot; the value is in watching movement as models like GPT-6.1 Terra roll out.
How reconnAI tracks GPT-6.1 Terra
reconnAI monitors how the leading AI models — ChatGPT, Claude, Gemini, Perplexity, Copilot and Google AI Overview — answer questions across regions, and re-baselines that tracking whenever a platform ships something that changes behaviour. A new, cheaper model reaching both professional and developer surfaces is exactly that kind of moment.
It joins a busy run of shifts we have followed lately, from the GPT-6 Sol and Luna default and Claude Haiku 5.5 to the new plugin directories and virtual try-on in ChatGPT. If you want to see how these models represent your brand when people ask about your category, get in touch with our team, explore how AI visibility tracking works, or read our overview of ChatGPT advertising for where paid placement is heading.
GPT-6.1 Terra: your questions answered
What did OpenAI launch? On 26 October 2026, OpenAI began rolling out GPT-6.1 Terra in ChatGPT Work and Codex and in the OpenAI API as gpt-6.1-terra, a broad model for visual reasoning, computer use and writing.
How is Terra different from Sol? Sol is the agentic specialist built for coding and computer use; Terra is a well-rounded, keenly priced model aimed at general professional and developer tasks, with a strong visual-reasoning focus.
Why does a cheaper model matter for AI visibility? Lower API pricing means more products can run a capable model, so more answers about your category get generated — and more moments arise where a model decides whether to name you.
Can I use GPT-6.1 Terra in the ChatGPT Chat tab? Not at launch. In ChatGPT it is available in Work and Codex, not the Chat tab, and separately in the OpenAI API.
How do I track my visibility across models? Baseline the questions your customers ask, run them across ChatGPT, Gemini, Claude and Perplexity, and watch how the answers change over time — which is what reconnAI automates.
About reconnAI
reconnAI tracks how the major AI models represent topics and sources across ChatGPT, Claude, Gemini, Perplexity, Copilot and Google AI Overview — across multiple regions. We monitor how those models answer and how they change over time, so you can stay ahead of shifts in the AI landscape.
Last updated: 28 October 2026.