🔍 Read the full analysis: GPT-6.1 Terra: The Affordable Model That Could Fit The Task on ThorstenMeyerAI.com
Get business pricing on tech for your team
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
OpenAI launched GPT-6.1 Terra on October 26, 2026, pricing it at $1 per million input tokens and $5 per million output tokens. Artificial Analysis measured its cost at $0.05 to $0.40 per benchmark task across effort settings, with higher settings improving scores but raising costs; the results do not make Terra the top-performing model.
OpenAI released GPT-6.1 Terra on October 26, 2026, adding a lower-priced model between its Luna and Sol/Astra offerings. Its API costs $1 per million input tokens and $5 per million output tokens, while independent benchmarking by Artificial Analysis found that the cost to complete its evaluation tasks ranged from about $0.05 to $0.40, depending on the model’s effort setting.
OpenAI describes Terra as a fast, efficient model for communication and computer-based work, including high-volume tasks such as processing invoices. It is available through the API under the identifier gpt-6.1-terra and to eligible Plus, Pro, Enterprise and Edu users in ChatGPT Work and Codex. OpenAI says its API token prices are 50% lower than those for GPT-5.6 Terra and GPT-6.1 Sol. Cached input is priced at $0.05 per million tokens.
Artificial Analysis’s launch-day evaluation gave Terra a 48.1 Intelligence Index score at maximum effort. That is below GPT-6.1 Sol’s 51.8 and Claude Opus 5.5’s 57.6, but above GPT-6 Luna’s 38.1. The firm calculated average task costs of $0.40 for Terra at maximum effort, $0.72 for Sol and $5.98 for Opus 5.5. Those figures describe the benchmark’s task set, not a guaranteed cost for every real-world request.
Effort setting changes both results and spending. Artificial Analysis measured six settings, with scores ranging from 33.5 in non-reasoning mode to 48.1 at maximum effort; reported average task costs ranged from $0.055 at low effort to $0.40 at maximum. The source’s summary says low and medium settings sit on its cost-efficiency frontier, while higher settings cost more for their additional score gains. It also reports that non-reasoning mode costs almost twice as much per task as low effort while scoring lower.
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 Change the Comparison
Terra’s launch highlights why token prices do not alone determine the cost of using a model. A cheaper model can produce more tokens or require more work to finish a task; a higher-priced model may use fewer. Artificial Analysis says GPT-6 Luna’s token price is ten times lower than Terra’s, yet Terra at low effort costs less per task in its evaluation because it generates fewer tokens. That is a result from the benchmark, not a universal rule for all prompts.
For organizations handling repeatable, high-volume work, the findings support testing models against the cost per completed task, not just the listed price per million tokens. Effort settings provide another control: routine work may suit a lower setting, while more complex tasks may call for greater reasoning or a more capable model. The benchmark does not establish that Terra will match larger models on every business workflow.
The results also show a performance trade-off. At maximum effort, Terra remains below several competitors in Artificial Analysis’s index. The same firm lists open-weight MiMo-V2.6-Pro at a score of 46.3 and $0.13 per task, lower cost and a higher score than Terra’s high and extra-high settings according to the source. Comparisons depend on model configuration and benchmark methodology, so buyers should validate them on their own workloads.
Products Worth Considering
LG UltraGear 32GX850A-B 32-Inch 4K UHD Glossy OLED Gaming Monitor
Samsung 27 Odyssey OLED G5 (G50SF) QHD QD-OLED Gaming Monitor
Samsung 32-Inch Odyssey OLED G8 (G80SH) 4K Gaming Monitor
As an affiliate, we earn on qualifying purchases.
How Terra Fits OpenAI’s Lineup
OpenAI positions Terra between GPT-6 Luna and the larger Sol and Astra models. Its launch materials present Terra for communication and computer-use tasks, while stating that Sol and Astra remain better choices for the most complex coding work. The company also says Terra completes tasks more than three times faster than Astra and Sol, but Artificial Analysis had not published speed measurements for Terra at the time of the source report. The speed comparison is therefore an OpenAI claim, not an independently confirmed result in that evaluation.
OpenAI supplied its own benchmark results as well. On Agents’ Last Exam, the company reported Terra at medium effort scoring 57.6% and costing $0.29 per task, compared with GPT-5.6 Terra’s 42.6% and $0.84. At maximum effort, OpenAI’s chart showed Terra scoring 60.0% on that evaluation, below Claude Opus 5.5’s 63.2%, while costing $1.01 versus $8.82. These are company-provided measurements and should be distinguished from Artificial Analysis’s independent index.
OpenAI also reported improved results on its internal error-inducing prompts: the error rate fell from 53.6% to 16.7% at low effort and from 22.3% to 4.3% at maximum effort, relative to GPT-5.6 Terra. On Terminal-Bench 4.0, it reported a 50.3% solve rate at maximum effort, compared with 21.5% for the predecessor’s published result. These results use different tests and do not by themselves establish performance across all coding or factual tasks.
Limits of the Benchmark Evidence
The published figures do not establish how Terra will perform or cost on every customer’s workload. Artificial Analysis’s task costs are averages across its own evaluation set, and OpenAI’s results come from company-selected tests. Independent speed measurements were not available in the cited evaluation when the source report was prepared, so OpenAI’s faster-completion claim could not be checked against that source.
The source material does not provide the full results for every one of Terra’s six effort settings, including all intermediate costs and scores. It also does not specify how actual API bills will vary with prompt length, repeated use, tool calls or particular task designs. OpenAI’s cybersecurity classification is likewise its own assessment under its Preparedness Framework, rather than an independent certification described in the source.
What Users Should Test Next
Businesses considering Terra can compare it with their current models on representative tasks, tracking completion quality, failure rates, latency and total cost per finished task. Testing low and medium effort first may help identify whether those settings meet the required quality bar before teams pay for more reasoning. For difficult coding or other demanding work, OpenAI itself points users toward Sol or Astra.
Further evidence will depend on broader use and additional evaluations, including independent speed results and performance on real-world workflows. Until those are available, the launch establishes Terra’s pricing and initial benchmark results, but not a universal cost advantage or a single best setting for every user.
Key Questions
How much does GPT-6.1 Terra cost?
OpenAI lists API prices of $1 per million input tokens and $5 per million output tokens, with cached input at $0.05 per million tokens. Artificial Analysis reported average benchmark task costs from about $0.055 at low effort to $0.40 at maximum effort; those are evaluation averages, not fixed prices per task.
Where can users access Terra?
The source says Terra is available in the API as gpt-6.1-terra and to Plus, Pro, Enterprise and Edu users in ChatGPT Work and Codex.
Is Terra OpenAI’s highest-performing model?
No. Artificial Analysis scored Terra at 48.1 at maximum effort, below GPT-6.1 Sol at 51.8 and Claude Opus 5.5 at 57.6 in its reported index. The source says OpenAI recommends Sol and Astra for its most complex coding tasks.
Which Terra effort setting is cheapest?
In Artificial Analysis’s measurements, low effort had the lowest reported task cost, at $0.055 on average. The source also reports that non-reasoning mode cost almost twice as much per task as low effort while scoring lower. Actual costs can vary with the work being done.
Are the performance and speed claims independently verified?
Artificial Analysis published an independent benchmark evaluation, but several figures in the source—including OpenAI’s internal error-rate results and speed comparison—are OpenAI’s own claims. The source says Artificial Analysis had not published speed measurements for Terra when the report was prepared.
Source: ThorstenMeyerAI.com
