🔍 Read the full analysis: How GPT-6.1 Terra Could Lower The Cost Of Getting Work Done on ThorstenMeyerAI.com
Prime made for students and young adults
- Fast, free delivery for dorm and study essentials
- Prime Video and Amazon Music included
- Member-only deals
TL;DR
OpenAI released GPT-6.1 Terra on October 26, 2026, pricing API use at $1 per million input tokens and $5 per million output tokens. Artificial Analysis measured its per-task cost at $0.055 on low effort and $0.40 at maximum effort, while finding the cheaper low and medium settings offer better cost efficiency than higher settings. Terra is not the top-scoring model, and its results vary by task and effort level.
OpenAI released GPT-6.1 Terra on October 26, pricing it at $1 per million input tokens and $5 per million output tokens—half the listed rates for GPT-5.6 Terra. The model is aimed at routine work and computer-based tasks, but independent testing suggests businesses should compare the cost per completed task, not token prices alone: Terra’s measured task cost ranged from $0.055 at low effort to $0.40 at maximum effort.
OpenAI describes Terra as a fast, efficient model for communication and computer-based work, including processing invoices and entering information into accounting software. It is available through the API as gpt-6.1-terra and in ChatGPT Work and Codex for Plus, Pro, Enterprise and Edu users. The company says the same allowance can cover substantially more tasks with Terra than with its Sol or Astra models.
Artificial Analysis, a benchmarking firm, gave Terra an Intelligence Index score of 48.1 at maximum effort and estimated its average cost across the index’s evaluation tasks at $0.40. GPT-6.1 Sol scored 51.8 at $0.72 per task; Claude Opus 5.5 scored 57.6 at $5.98. Those results place Terra below the highest-scoring systems, while showing a lower measured task cost than those two comparators. They do not establish that Terra is the least expensive option: the open-weight MiMo-V2.6-Pro scored 46.3 for $0.13 per task in the same table.
Terra offers six effort settings, trading additional reasoning for higher measured cost. Artificial Analysis found that low effort cost $0.055 per task and scored 36.2, while maximum effort cost $0.40 and scored 48.1. The firm’s summary places the low and medium settings on its cost-efficiency frontier; higher settings cost more without a better score-to-cost balance by that measure. Its evaluation also found the non-reasoning setting was not the cheapest: it cost $0.107 per task and scored 33.5.
Infographic 01 · The effort curve
One model, six settings, a 7-fold cost range.
GPT-6.1 Terra at each effort level, as measured by Artificial Analysis. Low and medium sit on its cost-efficiency frontier: no tested model is both cheaper and better.
Score · Intelligence Index
independent higher is better
Cost per finished task
note the last step shorter is cheaper
All six settings
| Effort | Index | Cost / task | Output tokens / task | Hallucination rate |
|---|---|---|---|---|
| Non-reasoning | Index 33.5 | Cost $0.107 | Tokens 4,737 | Hallucination 72% |
| Low | Index 36.2 | Cost $0.055 | Tokens 2,681 | Hallucination 46% |
| Medium | Index 41.3 | Cost $0.095 | Tokens 5,835 | Hallucination 44% |
| High | Index 43.9 | Cost $0.134 | Tokens 9,027 | Hallucination 43% |
| Xhigh | Index 45.4 | Cost $0.194 | Tokens 14,551 | Hallucination 42% |
| Max | Index 48.1 | Cost $0.403 | Tokens 35,096 | Hallucination 38% |
Hallucination rate in AA-Omniscience: how often the model answers anyway when it does not know. Lower is better.
The first steps up are cheap. The last one doubles the bill.Assessment: Thorsten Meyer
Sources: Artificial Analysis model pages (retrieved 26 Oct 2026); Artificial Analysis launch article (26 Oct 2026). Cost per task is the average across the index’s evaluation tasks.
Powered by Thorsten Meyer AI
Task Costs Shape Terra’s Value
The release matters most to organizations that run large volumes of repeatable tasks. A lower token rate can reduce bills, but token use differs by model and task: Artificial Analysis reported that GPT-6 Luna, despite a lower per-token price, cost $0.07 per evaluated task at its maximum setting, compared with Terra’s $0.055 at low effort. That comparison uses different settings, so it does not show Terra is broadly cheaper than Luna. It does illustrate why the bill for a finished task can tell a different story from the rate per token.
The effort controls also give companies a way to match spending to the work. Routine extraction or classification may not need maximum reasoning, while complex coding or computer-use tasks could benefit from a stronger setting or a larger model. This is a potential budgeting advantage, not a guarantee of lower total costs: the benchmarks average across defined tests, and actual spending will depend on task mix, retries, integrations and the accuracy required.
OpenAI’s own results point to possible gains in some areas, but should be treated as company-reported measurements. For example, OpenAI said Terra at medium effort scored 57.6% on Agents’ Last Exam for $0.29 per task, compared with GPT-5.6 Terra’s 42.6% and $0.84. It also reported a 50.3% result on Terminal-Bench 4.0 at maximum effort. Independent testing and customer workloads will be needed to show how those results translate into routine use.
Products Worth Considering
Samsung 27 Odyssey OLED G5 (G50SF) QHD QD-OLED Gaming Monitor
Samsung 27-Inch Odyssey QD-OLED G8 4K 240Hz Gaming Monitor
Samsung 27-inch Odyssey OLED G6 QD-OLED QHD Gaming Monitor
As an affiliate, we earn on qualifying purchases.
Where Terra Fits in OpenAI’s Lineup
OpenAI places Terra between its smaller Luna model and its larger Sol and Astra models. The company’s listed API rates for Terra are $1 per million input tokens and $5 per million output tokens, with cached input priced at $0.05 per million tokens. The source report lists GPT-5.6 Terra at $2 and $12, respectively, and GPT-6.1 Sol at $2 and $10. These are token prices; Artificial Analysis’s task-cost estimates use the spending observed across its own evaluation set.
Artificial Analysis said its Intelligence Index combines ten evaluations, covering areas such as office work, terminal tasks, science and hallucination. Its launch-day model pages list Terra below Claude Opus 5.5, Kimi K3.1 Pro and Gemini 4 Argon in overall score. The same source reported a Terra score of 48 in its Coding Agent Index, compared with 54 for Sol and 31 for Luna. Rankings and costs apply to those tests and settings; they are not universal measures of performance in every workplace.
OpenAI also reported that Terra completes tasks more than three times faster than Astra and Sol, and described its cybersecurity classification under the company’s Preparedness Framework. Those statements are from OpenAI. Artificial Analysis had not published speed measurements for Terra at the time cited in the source report, so the speed claim was not independently verified there.
Questions Before Workplace Rollout
The published figures do not show how Terra will perform on each company’s own documents, software or error-sensitive workflows. Artificial Analysis’s task costs are averages across its evaluation tasks, and OpenAI’s benchmark figures are company-reported. The source material does not provide enough detail to calculate a typical customer’s monthly savings or compare the cost of completing an identical set of real business tasks across all models.
It also remains unclear how often users will need to move from low or medium effort to a more expensive setting, or how error rates and review requirements affect total costs in production. OpenAI reported lower error rates on its internal prompts and improved results on selected benchmarks, but those tests do not by themselves establish accuracy across workplace tasks. Independent speed measurements were not available in the cited material.
Measure Costs on Real Workloads
Organizations considering Terra can compare it with their current model on a representative batch of tasks, recording the cost per accepted result, completion time and need for human correction at each effort level. Starting with low or medium effort may help establish a baseline, but the right setting depends on the consequences of mistakes and the complexity of the work.
Further independent evaluations, customer deployment data and any new benchmark measurements will clarify whether Terra’s reported cost advantages hold beyond the tests available at launch. OpenAI’s own guidance also leaves room for Sol and Astra on the hardest coding tasks. For now, Terra’s clearest case is a lower-priced option whose value depends on matching model, effort and task—not simply choosing the lowest token rate.
Key Questions
What is GPT-6.1 Terra?
GPT-6.1 Terra is an OpenAI model released on October 26, 2026, for communication and computer-based work. It is available through the API, ChatGPT Work and Codex, according to OpenAI.
How much does GPT-6.1 Terra cost?
OpenAI lists API pricing at $1 per million input tokens and $5 per million output tokens, with cached input at $0.05 per million tokens. Artificial Analysis estimated average cost per evaluation task from $0.055 at low effort to $0.40 at maximum effort.
Is Terra cheaper than GPT-6.1 Sol?
Terra has lower listed API token prices than Sol in the source material. Artificial Analysis also estimated $0.40 per task for Terra at maximum effort and $0.72 for Sol at maximum effort across its evaluation set. Results on a particular workload may differ.
Which Terra effort setting should businesses use?
Artificial Analysis’s results favor low and medium effort on its cost-efficiency frontier. Businesses should test settings on their own work and weigh cost, accuracy and review needs; the source does not establish one setting as best for every task.
Is GPT-6.1 Terra OpenAI’s most capable model?
No. The source places Terra below OpenAI’s Sol and Astra models in the lineup, and Artificial Analysis recorded lower overall scores for Terra than for several other tested models. OpenAI said Sol and Astra remain better choices for its most complex coding tasks.
Source: ThorstenMeyerAI.com
