How fast is GPT-6.1 Terra compared to other models?

At 16.1 seconds per task, it is the fastest OpenAI model on the benchmark and the third fastest overall.

How much does GPT-6.1 Terra cost per task?

It costs $0.042 per task, about half of GPT-6.1 Sol and a tenth of GPT-6 Astra.

What areas did GPT-6.1 Terra perform best and worst in?

It scored 100% on transaction classification (13 of 13) but only 53.8% on accounts payable (7 of 13), tied for last.

Large Language ModelsAI Business & IndustryYahoo Finance AI1m agoOct 27, 2026, 06:01 JST

OpenAI's GPT-6.1 Terra ties for 18th of 22 on accounting benchmark

Send on LINE

3 Key Points

What happened
OpenAI's GPT-6.1 Terra scored 78.2% on the DualEntry Accounting AI Benchmark, tying for 18th of 22 models, with accounts payable at 53.8% (7 of 13), tied for last.
Why it matters
The result suggests speed and low cost ($0.042 per task) may not translate into accuracy on routine finance work, with three cheaper models scoring higher.

WHO IT HITSFinance teams evaluating AI for accounting tasks may need to test models on their own workflows rather than relying on speed or general benchmark scores, as Terra's weak accounts payable score shows.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Get these free every morning

Context & Analysis

The DualEntry Accounting AI Benchmark tests models on 101 synthetic accounting cases across eight areas, from transaction classification to month-end close. Each model completes the full benchmark three times from an identical test company with the same tools and instructions, and answers are checked against reference answers written in advance using fixed scoring rules rather than an AI grader. No customer data is used.

GPT-6.1 Terra's performance was uneven: it was competitive in transaction classification, journal entries (92.3%) and bank reconciliation (83.3%, behind only Kimi K3.1 Pro and Muse Wonder), but struggled in accounts payable and accounts receivable (58.3%). Its consistency score was 65.3%, second lowest after GPT-6 Luna, and it scored 71.4% on hard tasks but only 75.7% on easy ones, the second-lowest easy-task score.

Compared with peers, Terra trailed Google's Gemini 4 Argon at 91.1%, Zhipu AI's GLM-5.3 Flash at 88.1% and SpaceXAI's Grok 4.7 at 88.1%, as well as OpenAI's own GPT-6.1 Sol at 81.2%. On cost, Terra's $0.042 per task is about half of GPT-6.1 Sol and a tenth of GPT-6 Astra, but three cheaper models scored higher: DeepSeek V4.1 Flash, GLM-5.3 Flash and GLM-5.3.

FAQ

How fast is GPT-6.1 Terra compared to other models?: At 16.1 seconds per task, it is the fastest OpenAI model on the benchmark and the third fastest overall.
How much does GPT-6.1 Terra cost per task?: It costs $0.042 per task, about half of GPT-6.1 Sol and a tenth of GPT-6 Astra.
What areas did GPT-6.1 Terra perform best and worst in?: It scored 100% on transaction classification (13 of 13) but only 53.8% on accounts payable (7 of 13), tied for last.

Yahoo Finance AIRead Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.