Today · launch, October 26

OpenAI launches GPT-6.1 Terra; figures are self-reported

OpenAI launches GPT-6.1 Terra as a mid-priced tier for computer use; every capability figure is OpenAI's own

Our AI analysis, unconfirmed: OpenAI is pitching computer use at a mid-priced tier, and only OpenAI's figures show whether a task now costs less Check back when when a third party measures cost per task on the same computer-use tasks

Concept illustration: a large billboard on posts above a rooftop is lettered OPENAI in black beside an interlaced knot-shaped symbol, with a pale pink circle behind its top-right corner. On it, two open laptops sit side by side, a dark grey one on the left labelled GPT-6.1 SOL and a red one on the right labelled GPT-6.1 TERRA; each screen shows two overlapping windows, one with grey text lines and a button with a mouse pointer on it and one with a speech bubble, and there are two blank round discs below the left label and one below the right.

Why it matters: OpenAI is pitching computer use at a mid-priced tier. Teams choosing an agent model can't yet tell whether a task really costs less, because only OpenAI has measured it.

OpenAI released GPT-6.1 Terra on October 26 at US$1 per million input tokens and US$5 per million output tokens, writing that its standard API prices are 50% lower than those of GPT-5.6 Terra and GPT-6.1 Sol (OpenAI, 2026-10-26). Microsoft's Foundry list put GPT-5.6 Terra at US$2.50 input and US$15 output (Microsoft Azure, 2026-07-09), but Foundry's list price can differ from OpenAI's direct price, and the launch page doesn't say which list price the 50% is measured against. OpenAI claims a ~9x lower cost per task than Claude Opus 5.5 on its own Agents' Last Exam, though multiplying its two factors gives more than ninefold. An OpenAI employee called Terra "a nice midrange option if you want something cheaper than Sol but a lot smarter than Luna," a comment Simon Willison relayed (Simon Willison, 2026-10-26).

Not proven yet: cost per task has been measured only by OpenAI How we checked →

Judgment update: Our reading: whether enterprises switch tiers turns on cost per task, not the per-token list price. Both factors behind the cost-per-task figure have so far been measured only by OpenAI. If a third party measures Terra on the same computer-use tasks and finds its cost per task no lower than comparable rivals', Terra is cheaper only per token, not per task. It becomes a price war only if Anthropic's Sonnet tier or Google's equivalent cuts prices within two weeks.

Source and checks

This item ran in the brief of Oct 27, 2026, alongside 12 other items. How we checked it, the quote checks and what was still missing that day are on that issue's How we checked page.

Cite this item

SecondSource (2026). "OpenAI launches GPT-6.1 Terra; figures are self-reported." https://secondsource.io/en/news/2026-10-27-2/. Accessed access date.