![]()
Google’s newly released Gemini 4 Neon got 91.1% of accounting tasks right on the DualEntry Accounting AI Benchmark, tying Google’s Gemini 4 Argon for the highest score among 29 leading AI models tested.
Neon beat Argon on bank reconciliation (91.7% vs 75%) and on hard tasks (85.7% vs 82.1%). Argon was slightly more consistent (86.1% vs 85.1%) and cheaper per task.
Google’s Gemini 4 Carbon is tied for third at 88.1% with Zhipu AI’s GLM-5.3 Flash and SpaceXAI’s Grok 4.7. That puts Google’s models first, second and tied for third on the benchmark, the strongest showing by any provider.
Key findings:
- Gemini 4 Neon scored 91.1% overall, tied for first of 29 with Gemini 4 Argon
- Google’s models are first, second and tied for third: Neon and Argon at 91.1%, then Gemini 4 Carbon at 88.1%, level with GLM-5.3 Flash and Grok 4.7
- It leads Anthropic’s best model, Claude Fable 5.1 (85.1%), by 6.0 points, and OpenAI’s best, GPT-6.2 Sol (84.2%), by 6.9 points
- It scored 100% on transaction classification and financial reporting, and 91.7% on accounts receivable and bank reconciliation
- It scored 85.7% on hard tasks (24 of 28), ahead of Argon’s 82.1%, and 97.2% on medium tasks (35 of 36)
- Its consistency score was 85.1%, compared with 86.1% for Gemini 4 Argon
- It costs $0.174 per task ($52.76 for a full benchmark run), about 50% more than Gemini 4 Argon ($0.115), and averages 24.3 seconds per task
- On bank reconciliation it scored 91.7%, 16.7 points ahead of Argon
Gemini 4 Neon was not the strongest model in every category. Its weakest result came in month-end close at 66.7% (8 of 12), where Moonshot AI’s Kimi K3 led at 91.7% and Argon scored 83.3%.
“Google now has two models tied at the top of our benchmark, and both get nine in ten accounting tasks right,” said Benedict Dohmen, co-founder of DualEntry. “For finance teams the question is no longer whether AI can do the work, but whether it gets the same answer every time, and that is what separates the leaders.”
The DualEntry Accounting AI Benchmark tests models on 101 synthetic accounting cases across eight areas: transaction classification, journal entries, accounts payable, accounts receivable, bank reconciliation, financial reporting, month-end close, and accounting knowledge.
Each model completes the full benchmark three times from an identical test company with the same tools and instructions. Answers are checked against reference answers written in advance using fixed scoring rules rather than an AI grader. No customer data is used.
Full results for all 29 models, including accuracy, consistency, cost, response time, and performance by accounting area, are available at dualentry.com/benchmark.
About DualEntry
DualEntry is the agentic, AI-native ERP for mid-market finance teams. It brings the general ledger, accounts payable and receivable, revenue recognition, multi-entity consolidation, and reporting into one platform, with AI agents that take on accounting work from data entry through close. Founded in 2024 and based in New York, DualEntry raised a $90 million Series A in 2025 from Lightspeed Venture Partners, Khosla Ventures, and GV. Learn more at DualEntry.com.
View source version on businesswire.com: https://www.businesswire.com/news/home/20261119444436/en/
Media gallery