Kaggle Can AI Close the Books (FP&A Reasoning): leaderboard
Kaggle Benchmarks corporate finance-close evaluation (marcopolofpa/can-ai-close-the-books-fp-and-a-reasoning): 28 deterministic synthetic FP&A reasoning tasks covering penny-reconciled variance bridges, accruals and cut-off adjustments, three-statement linkage, and valid/invalid reviewer claims. Exact pass rate (0-100%); higher is better.
Metric: Pass Rate (%). Source: www.kaggle.com. Saturation forecast: Around February 2027. 40 models tracked. Updated 21 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 100 | #276 |
| 2 | Gemini 3.1 Pro (Preview) | 100 | #99 |
| 3 | GPT-5.5 | 100 | #54 |
| 4 | GPT-5.6 Sol | 100 | #37 |
| 5 | Gemini 3 Flash (Preview) | 100 | #152 |
| 6 | Gemma 4 31B | 100 | #382 |
| 7 | GLM-5 | 100 | #269 |
| 8 | Claude Opus 5 | 100 | #26 |
| 9 | Gemini 3.8 Flash | 100 | #43 |
| 10 | Claude Opus 5.5 | 100 | #9 |
| 11 | GPT-5.6 Luna | 100 | #131 |
| 12 | Gemma 4 26B A4B | 100 | #590 |
| 13 | Claude Sonnet 5 | 100 | #118 |
| 14 | GPT-6 Luna | 100 | #105 |
| 15 | Gemini 3.6 Flash | 100 | #81 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=kaggle-can-ai-close-the-books-fp-a-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-21.