🔍 Read the full analysis: GPT-6.1 Terra May Be The Practical Pick For Lower-Cost AI on ThorstenMeyerAI.com
Get hardware and tech essentials delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
OpenAI released 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 task costs from about $0.05 to $0.40 across its effort settings, but Terra scored below several stronger models and was not the cheapest option in every comparison.
OpenAI released GPT-6.1 Terra on October 26, positioning it as a lower-cost model for routine workplace and computer-based tasks. Its API price is $1 per million input tokens and $5 per million output tokens, while independent testing by Artificial Analysis found that cost per evaluated task ranged from about $0.05 to $0.40 depending on the effort setting.
Terra is available in the API as gpt-6.1-terra and, according to OpenAI, to Plus, Pro, Enterprise and Edu users in ChatGPT Work and Codex. OpenAI describes it as a fast, efficient model for communication and computer tasks, including work such as reading invoices and entering their details into accounting software. The company says the same ChatGPT allowance can cover substantially more tasks with Terra than with Sol or Astra.
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. The source article reports those rates are half the price of GPT-5.6 Terra and GPT-6.1 Sol. Those are token prices, not a guarantee of how much any particular task will cost: the number of tokens a model uses also affects the bill.
Artificial Analysis’s launch-day evaluation gave Terra a 48.1 Intelligence Index score at maximum effort, at an average cost of $0.40 per task across its index evaluations. The firm reported GPT-6.1 Sol at 51.8 and $0.72 per task, and Claude Opus 5.5 at 57.6 and $5.98. These are results from the firm’s benchmark and task-cost methodology, not prices for every real-world use.
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 Vary With Reasoning Effort
The release puts attention on a distinction that can matter to companies running AI at scale: price per token is not the same as cost per completed task. A cheaper model may use more tokens, while a more capable model may finish a task with fewer attempts. Artificial Analysis’s measurements account for spending across its benchmark tasks, offering a different comparison from the models’ listed token rates.
In its evaluation, Terra’s six effort settings ranged from an Intelligence Index score of 33.5 to 48.1, while average task cost rose from about $0.05 at low effort to $0.40 at maximum effort. The source reports that maximum effort cost 7.4 times as much as low effort for 12 additional score points. That makes the setting a practical cost-and-performance decision, rather than a simple choice between the model’s cheapest and most expensive price label.
The results do not establish Terra as the best or cheapest model for every workload. Artificial Analysis ranked several models above it on its index, and the source reports Xiaomi’s open-weight MiMo-V2.6-Pro scored 46.3 for $0.13 per task. For users and businesses, Terra’s relevance is that it offers another option to test for routine, repeated work—not that it removes the need to check output quality, reliability or total operating costs.
Products Worth Considering
Logitech MX Master 4 for Mac, Ergonomic Bluetooth Mouse with Haptic Feedback, Ultra-Fast Scrolling, USB-C Charging, for MacOS, iPadOS, Windo
Bose QuietComfort Wireless Bluetooth Headphones with Active Noise Cancellation, Mic, USB‑C Charging, Deep Bass, 24‑Hour Battery – White Smok
Logitech MX Mechanical Wireless Illuminated Keyboard, Tactile Quiet, Graphite
As an affiliate, we earn on qualifying purchases.
OpenAI’s Positioning and Test Results
OpenAI places Terra between its smaller GPT-6 Luna and larger Sol and Astra models. The company’s launch information describes Terra as designed for common workday tasks rather than the most demanding problems. OpenAI also says the model completes tasks more than three times faster than Astra and Sol; that is a company claim, and Artificial Analysis had not published speed measurements for Terra at the time described in the source.
OpenAI supplied its own benchmark results as well as pricing. On Agents’ Last Exam, the company reports Terra at medium effort scored 57.6%, compared with 42.6% for GPT-5.6 Terra, at $0.29 rather than $0.84 per task. On Terminal-Bench 4.0, OpenAI reports a 50.3% score at maximum effort, against 21.5% for GPT-5.6 Terra’s published result. The source also says OpenAI identifies Sol and Astra as better choices for its most complex coding tasks.
Artificial Analysis’s index combines ten evaluations, including agentic office work, terminal tasks, science and hallucination testing. Its figures describe results across that specific set. The source reports Terra guessed on 38% of questions it could not answer in the firm’s hallucination test, compared with 88% for GPT-5.6 Terra. The comparison is tied to that test and should not be read as a general rate for all user interactions.
Real-World Costs Need Testing
The benchmark figures do not show what Terra will cost for every company’s actual workload. Task length, input data, retries, tool use and selected effort can change usage and results. Artificial Analysis’s cost-per-task figure is an average across its evaluation tasks; it is not a quoted fixed charge for a specific business process.
Some comparisons also come from different sources. OpenAI’s benchmark claims are company-reported, while Artificial Analysis provides an independent evaluation using its own tests and methodology. The source does not provide independent speed measurements, and it does not establish how Terra performs across all languages, industries or production settings. OpenAI’s statement that Terra meets a cybersecurity threshold is likewise the company’s own classification under its Preparedness Framework.
The supplied material gives a partial effort-setting table and selected figures, but not a complete set of results for every setting in the article’s summary. The full underlying benchmark data, real-world adoption and longer-term reliability are not established here. Buyers should treat the published results as evidence to test against their own tasks, not as a universal ranking.
Measure Costs on Actual Workloads
The next useful test for businesses is to run Terra on representative tasks and compare cost per successfully completed task, accuracy and review time with the models they already use. Teams can begin with low or medium effort for routine tasks, then test whether higher effort or a larger model improves outcomes enough to justify the additional expense. OpenAI’s stated availability in the API, ChatGPT Work and Codex gives users several routes for evaluation.
Further comparisons will depend on additional independent measurements, including speed results and performance across more real-world tasks. Until those are available, Terra’s launch establishes its listed prices and initial benchmark results, but not a settled verdict on which model is most economical for a particular user or organization.
Key Questions
How much does GPT-6.1 Terra cost?
OpenAI lists $1 per million input tokens and $5 per million output tokens for API use, with cached input at $0.05 per million tokens. Actual task costs also depend on token use and the task itself.
Where is GPT-6.1 Terra available?
The source says Terra is available through the API as gpt-6.1-terra and to Plus, Pro, Enterprise and Edu users in ChatGPT Work and Codex.
Is Terra the cheapest model in the independent comparison?
No. Artificial Analysis reported a $0.40 average task cost for Terra at maximum effort, while its comparison included lower-cost models, including MiMo-V2.6-Pro at $0.13 per task. Results depend on the benchmark and setting.
Which Terra effort setting should users choose?
The source’s findings suggest comparing low or medium effort for routine work and testing higher effort where accuracy or task difficulty warrants it. There is no single setting shown to be best for every workload.
Does Terra outperform OpenAI’s larger models?
Not across the reported comparisons. Artificial Analysis scored Terra below Sol and Astra on its Intelligence Index, and OpenAI says Sol and Astra remain better choices for its most complex coding tasks.
Source: ThorstenMeyerAI.com
