DualEntry Accounting AI Benchmark
Models use accounting tools to look up accounts, reconcile balances, and draft entries for a test company. DualEntry checks their answers against fixed references.
DualEntry lists scores for 29 model versions. Each model runs all 101 cases three times. A case counts toward accuracy after two passes.
- Made by
- DualEntry
- Released
- October 2026
- Kind
- Finance
- Level
- University
- Scored
- Checked automatically
- Answers
- Agent task
- Measures
- Capability
- Subject
- Economics and business
- Language
- English
- Interaction
- Agentic
- Environment
- APIs and tools
- Refresh
- One fixed set
- Run by
- A host
- Occupations
- Business and financial operations
- In views
- Agents, Law and finance
- Strongest model tested
- GPT-6.2 Astra #1 on the capability index
- Last model added
- November 2026
- Scores refreshed
- November 20th, 2026
2 models share the top score of 91.1%
Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.
- 91.1
Gemini 4 Argon
- 91.1
Gemini 4 Neon
- 88.1 GLM-5.3-Flash
- 88.1 Grok 4.7
- 88.1
Gemini 4 Carbon
- 86.1 Kimi K3.1 Pro
- 85.1 Claude Fable 5.1
- 84.2 Kimi K3
- 84.2
DeepSeek V4.1 Flash
- 84.2 GPT-6.2 Sol
- 83.2 Grok 4.5
- 83.2
Gemini 3.8 Flash
- 83.2
Muse Wonder
- 83.2
Gemini 4 Hydrogen
- 83.2
Qwen4-Max
- 82.2 GLM-5.3
- 82.2 Claude Opus 5.5
- 82.2
DeepSeek V4.1 Flash 1017
- 81.2 GPT-6 Astra
- 81.2 GPT-6.1 Sol
- 81.2 Claude Fable 5.2
- 80.2 Claude Sonnet 5.5
- 80.2
Mistral Large 4
- 78.2 Mercury 3
- 78.2 GPT-6.1 Terra
- 78.2 GPT-6.2 Astra
- 76.2 GPT-6 Sol
- 76.2 Claude Haiku 5.5
- 66.3 GPT-6 Luna
- 1
Gemini 4 Argon 91.1%
- 2
Gemini 4 Neon 91.1%
- 3 GLM-5.3-Flash 88.1%
- 4 Grok 4.7 88.1%
- 5
Gemini 4 Carbon 88.1%
- 6 Kimi K3.1 Pro 86.1%
- 7 Claude Fable 5.1 85.1%
- 8 Kimi K3 84.2%
- 9
DeepSeek V4.1 Flash 84.2%
- 10 GPT-6.2 Sol 84.2%
- 11 Grok 4.5 83.2%
- 12
Gemini 3.8 Flash 83.2%
- 13
Muse Wonder 83.2%
- 14
Gemini 4 Hydrogen 83.2%
- 15
Qwen4-Max 83.2%
- 16 GLM-5.3 82.2%
- 17 Claude Opus 5.5 82.2%
- 18
DeepSeek V4.1 Flash 1017 82.2%
- 19 GPT-6 Astra 81.2%
- 20 GPT-6.1 Sol 81.2%
- 21 Claude Fable 5.2 81.2%
- 22 Claude Sonnet 5.5 80.2%
- 23
Mistral Large 4 80.2%
- 24 Mercury 3 78.2%
- 25 GPT-6.1 Terra 78.2%
- 26 GPT-6.2 Astra 78.2%
- 27 GPT-6 Sol 76.2%
- 28 Claude Haiku 5.5 76.2%
- 29 GPT-6 Luna 66.3%
Source: Official DualEntry Accounting AI Benchmark leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 91.1% | Sep 2026 | Official leaderboard | |
| 2 | | 91.1% | Nov 2026 | Official leaderboard | |
| 3 | GLM-5.3-Flash | 88.1% | Aug 2026 | Official leaderboard | |
| 4 | Grok 4.7 | 88.1% | Sep 2026 | Official leaderboard | |
| 5 | | 88.1% | Oct 2026 | Official leaderboard | |
| 6 | Kimi K3.1 Pro | 86.1% | Oct 2026 | Official leaderboard | |
| 7 | Claude Fable 5.1 | 85.1% | Sep 2026 | Official leaderboard | |
| 8 | Kimi K3 | 84.2% | Jul 2026 | Official leaderboard | |
| 9 | | 84.2% | Sep 2026 | Official leaderboard | |
| 10 | GPT-6.2 Sol | 84.2% | Nov 2026 | Official leaderboard | |
| 11 | Grok 4.5 | 83.2% | Jul 2026 | Official leaderboard | |
| 12 | | 83.2% | Sep 2026 | Official leaderboard | |
| 13 | | 83.2% | Oct 2026 | Official leaderboard | |
| 14 | | 83.2% | Nov 2026 | Official leaderboard | |
| 15 | | 83.2% | Nov 2026 | Official leaderboard | |
| 16 | GLM-5.3 | 82.2% | Aug 2026 | Official leaderboard | |
| 17 | Claude Opus 5.5 | 82.2% | Sep 2026 | Official leaderboard | |
| 18 | | 82.2% | Oct 2026 | Official leaderboard | |
| 19 | GPT-6 Astra | 81.2% | Sep 2026 | Official leaderboard | |
| 20 | GPT-6.1 Sol | 81.2% | Sep 2026 | Official leaderboard | |
| 21 | Claude Fable 5.2 | 81.2% | Nov 2026 | Official leaderboard | |
| 22 | Claude Sonnet 5.5 | 80.2% | Sep 2026 | Official leaderboard | |
| 23 | | 80.2% | Oct 2026 | Official leaderboard | |
| 24 | Mercury 3 | 78.2% | Oct 2026 | Official leaderboard | |
| 25 | GPT-6.1 Terra | 78.2% | Oct 2026 | Official leaderboard | |
| 26 | GPT-6.2 Astra | 78.2% | Nov 2026 | Official leaderboard | |
| 27 | GPT-6 Sol | 76.2% | Sep 2026 | Official leaderboard | |
| 28 | Claude Haiku 5.5 | 76.2% | Oct 2026 | Official leaderboard | |
| 29 | GPT-6 Luna | 66.3% | Sep 2026 | Official leaderboard |
Rankings by part
We read 10 parts from the benchmark’s own leaderboard in 2 groups: accounting area and difficulty. Each ranks models on one part of the test. Kimi K3 leads 2 of them, and 7 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
Transaction Classification
25 models we track have a score here.
- 1 Claude Fable 5.1 ↑6 100%
- 2
Gemini 3.8 Flash ↑10 100%
- 3 GPT-6 Astra ↑16 100%
- 4
DeepSeek V4.1 Flash ↑5 100%
- 5 Grok 4.7 ↓1 100%
- 6 Claude Opus 5.5 ↑11 100%
- 7 Claude Sonnet 5.5 ↑15 100%
- 8 GPT-6.1 Sol ↑12 100%
- 9
Gemini 4 Argon ↓8 100%
- 10 Claude Haiku 5.5 ↑18 100%
Journal Entries
25 models we track have a score here.
- 1
Gemini 3.8 Flash ↑11 100%
- 2 Grok 4.5 ↑9 92.3%
- 3 Kimi K3 ↑5 92.3%
- 4 GLM-5.3 ↑12 92.3%
- 5 GLM-5.3-Flash ↓2 92.3%
- 6
DeepSeek V4.1 Flash ↑3 92.3%
- 7 Grok 4.7 ↓3 92.3%
- 8 Claude Opus 5.5 ↑9 92.3%
- 9 GPT-6 Sol ↑18 92.3%
- 10
Gemini 4 Argon ↓9 92.3%
Accounts Payable
25 models we track have a score here.
- 1 Grok 4.5 ↑10 84.6%
- 2 GLM-5.3-Flash ↑1 84.6%
- 3 Claude Fable 5.1 ↑4 84.6%
- 4
Gemini 3.8 Flash ↑8 84.6%
- 5 GPT-6 Astra ↑14 84.6%
- 6 Grok 4.7 ↓2 84.6%
- 7 GPT-6.1 Sol ↑13 84.6%
- 8
Gemini 4 Argon ↓7 84.6%
- 9 Kimi K3.1 Pro ↓3 84.6%
- 10 Claude Fable 5.2 ↑11 84.6%
Accounts Receivable
25 models we track have a score here.
- 1 Grok 4.7 ↑3 91.7%
- 2
Gemini 4 Argon ↓1 91.7%
- 3
Gemini 4 Neon ↓1 91.7%
- 4
Gemini 4 Carbon ↑1 83.3%
- 5 GLM-5.3-Flash ↓2 75%
- 6 GPT-6.1 Sol ↑14 75%
- 7
Mistral Large 4 ↑16 75%
- 8 Kimi K3.1 Pro ↓2 75%
- 9 GPT-6.2 Sol ↑1 75%
- 10 GLM-5.3 ↑6 66.7%
Bank Reconciliation
25 models we track have a score here.
- 1 Kimi K3.1 Pro ↑5 100%
- 2
Muse Wonder ↑11 91.7%
- 3
Gemini 4 Neon ↓1 91.7%
- 4 Grok 4.5 ↑7 83.3%
- 5 Claude Fable 5.1 ↑2 83.3%
- 6
DeepSeek V4.1 Flash ↑3 83.3%
- 7 Grok 4.7 ↓3 83.3%
- 8 Claude Fable 5.2 ↑13 83.3%
- 9 GPT-6.2 Sol ↑1 83.3%
- 10 GLM-5.3 ↑6 75%
Financial Reporting
25 models we track have a score here.
- 1 Kimi K3 ↑7 100%
- 2 GLM-5.3 ↑14 100%
- 3 GLM-5.3-Flash 100%
- 4 Claude Fable 5.1 ↑3 100%
- 5 Claude Sonnet 5.5 ↑17 100%
- 6
Gemini 4 Argon ↓5 100%
- 7
Muse Wonder ↑6 100%
- 8
Gemini 4 Hydrogen ↑6 100%
- 9
Gemini 4 Neon ↓7 100%
- 10 Grok 4.5 ↑1 92.3%
Month-End Close
25 models we track have a score here.
- 1 Kimi K3 ↑7 91.7%
- 2 GLM-5.3-Flash ↑1 83.3%
- 3
Gemini 4 Argon ↓2 83.3%
- 4
Gemini 4 Carbon ↑1 83.3%
- 5 Grok 4.5 ↑6 75%
- 6
DeepSeek V4.1 Flash ↑3 75%
- 7 Claude Fable 5.2 ↑14 75%
- 8 Claude Fable 5.1 ↓1 66.7%
- 9 Grok 4.7 ↓5 66.7%
- 10 Claude Opus 5.5 ↑7 66.7%
Easy
25 models we track have a score here.
- 1 GLM-5.3-Flash ↑2 94.6%
- 2 Claude Opus 5.5 ↑15 94.6%
- 3
Gemini 4 Argon ↓2 94.6%
- 4
Mistral Large 4 ↑19 94.6%
- 5 Kimi K3.1 Pro ↑1 94.6%
- 6 Kimi K3 ↑2 91.9%
- 7 Claude Fable 5.1 91.9%
- 8 Grok 4.5 ↑3 89.2%
- 9 GPT-6 Astra ↑10 89.2%
- 10
DeepSeek V4.1 Flash ↓1 89.2%
Medium
25 models we track have a score here.
- 1 GLM-5.3-Flash ↑2 97.2%
- 2
Gemini 4 Neon 97.2%
- 3
Gemini 4 Argon ↓2 94.4%
- 4 GLM-5.3 ↑12 91.7%
- 5 Grok 4.7 ↓1 91.7%
- 6 Claude Fable 5.1 ↑1 88.9%
- 7
Muse Wonder ↑6 88.9%
- 8 Grok 4.5 ↑3 86.1%
- 9 Kimi K3 ↓1 86.1%
- 10
DeepSeek V4.1 Flash ↓1 86.1%
Hard
25 models we track have a score here.
- 1
Gemini 4 Carbon ↑4 89.3%
- 2
Gemini 4 Neon 85.7%
- 3
Gemini 3.8 Flash ↑9 82.1%
- 4 Grok 4.7 82.1%
- 5
Gemini 4 Argon ↓4 82.1%
- 6
Gemini 4 Hydrogen ↑8 82.1%
- 7 GPT-6.2 Sol ↑3 78.6%
- 8
DeepSeek V4.1 Flash ↑1 75%
- 9 GPT-6.1 Sol ↑11 75%
- 10
Muse Wonder ↑3 75%
The best score, 91.1%, costs $0.11 a task
Runs on the cost frontier score higher than any run that's cheaper. GLM-5.3-Flash comes within 3 points of the best for $0.006 a task, 18 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: Official DualEntry Accounting AI Benchmark leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| 91.1% | $0.11 | ||||
| 91.1% | $0.17 | ||||
| GLM-5.3-Flash | 88.1% | $0.006 | |||
| Grok 4.7 | 88.1% | $0.13 | |||
| 88.1% | $0.08 | ||||
| Kimi K3.1 Pro | 86.1% | $0.08 | |||
| Claude Fable 5.1 | 85.1% | $1.18 | |||
| Kimi K3 | 84.2% | $0.08 | |||
| 84.2% | $0.003 | ||||
| GPT-6.2 Sol | 84.2% | $0.09 | |||
| Grok 4.5 | 83.2% | $0.09 | |||
| 83.2% | $0.06 | ||||
| 83.2% | $0.11 | ||||
| 83.2% | $0.06 | ||||
| 83.2% | |||||
| GLM-5.3 | 82.2% | $0.04 | |||
| Claude Opus 5.5 | 82.2% | $0.45 | |||
| 82.2% | |||||
| GPT-6 Astra | 81.2% | $0.49 | |||
| GPT-6.1 Sol | 81.2% | $0.09 | |||
| Claude Fable 5.2 | 81.2% | $1.13 | |||
| Claude Sonnet 5.5 | 80.2% | $0.22 | |||
| 80.2% | $0.04 | ||||
| Mercury 3 | 78.2% | ||||
| GPT-6.1 Terra | 78.2% | ||||
| GPT-6.2 Astra | 78.2% | $0.38 | |||
| GPT-6 Sol | 76.2% | $0.10 | |||
| Claude Haiku 5.5 | 76.2% | $0.01 | |||
| GPT-6 Luna | 66.3% | $0.005 |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official DualEntry Accounting AI Benchmark leaderboard
GLM-5.3-Flash on the official leaderboard passed 85% in August 2026
Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official DualEntry Accounting AI Benchmark leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Law, finance and medicine
- MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy.
Qwen 3.8 Max 80.4%
- HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
- HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
- FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy.
Ling-3.1-flash 81.7%
- KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. o1-preview 92.7%
- Harvey LAB-AA v1.1 In use Artificial Analysis's run of Harvey's Legal Agent Benchmark on Harvey's newer private set of 120 tasks across 24 practice areas. A deliverable with a material hallucination scores zero, and the headline is the share of judges who find every rubric criterion passed. GPT-6.2 Astra 15%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark
Catalog record 1 warning 1 pass
- Warning Task access We don’t know whether the tasks are public.
- Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
- Pass Independent runs Scores from the official leaderboard (29 entries).
- Pass Models scored 29 models we track.
- Pass Newest score The newest is from November 2026, for Gemini 4 Neon.
- Pass Frontier models 10 of the top 10 have a score, led by GPT-6.2 Astra (#1).
- Note Headroom Saturated. The best is 91.1%, past 85% since August 2026.
Leaderboard import 3 pass
- Pass Importer We read the board on dualentry.com: 29 rows for models we track at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.