Best AI for Spreadsheets and accounting (November 2026)
Data refreshed 1 Nov 2026
Gemini 4 Argon leads APEX-Accounting with 71.5 (Mean Score (% criteria passed)). Claude Fable 5.1 leads Excel Modeling Benchmark (EMB) with 76.7 (accuracy %). GPT 5.5 leads BlueFin with 49.6 (criteria-weighted score % (overall)). GPT 5.1 Pro leads Finch: Finance & Accounting Cowork Benchmark with 38.4 (workflow pass rate % (human eval)).
- Benchmarks in this use case
- 6
- With dated scores
- 4
- Models with scores
- 24
- Latest source date
- 1 Nov 2026
Spreadsheets and accounting picks
The leader on each benchmark, never combinedBest on APEX-Accounting
Gemini 4 ArgonGoogle
71.5 Mean Score (% criteria passed)
Official board, 1 Nov 2026
Best on Excel Modeling Benchmark (EMB)
Claude Fable 5.1Anthropic
76.7 accuracy %
Official board, 29 Oct 2026
Best on BlueFin
GPT 5.5OpenAI
49.6 criteria-weighted score % (overall)
Reported by the authors, May 2026
Best on Finch: Finance & Accounting Cowork Benchmark
GPT 5.1 ProOpenAI
38.4 workflow pass rate % (human eval)
Reported by the authors, Apr 2026
Spreadsheets and accounting leaderboard
APEX-Accounting Paper v2 (2026-07-30); leaderboard live, no update date shown
Mean Score (% criteria passed), higher is better. Top 5 of 10 models.
Updated 1 Nov 2026
- 1Gemini 4 Argon71.5
- 2Gemini 4 Carbon70.7
- 3Muse Wonder 1.067.1
- 4Fable 5.166.7
- 5Opus 5.565.5
| 1 | Gemini 4 Argon | 71.5 | High effort, Loop harness | Mercor APEX-Accounting leaderboard Official board, 1 Nov 2026 | |
|---|---|---|---|---|---|
| 2 | Gemini 4 Carbon | 70.7 | High effort, Loop harness | Mercor APEX-Accounting leaderboard Official board, 1 Nov 2026 | |
| 3 | Muse Wonder 1.0 | Meta | 67.1 | Max effort, Loop harness | Mercor APEX-Accounting leaderboard Official board, 1 Nov 2026 |
| 4 | Fable 5.1 | Anthropic | 66.7 | Max effort, Loop harness | Mercor APEX-Accounting leaderboard Official board, 1 Nov 2026 |
| 5 | Opus 5.5 | Anthropic | 65.5 | Max effort, Loop harness | Mercor APEX-Accounting leaderboard Official board, 1 Nov 2026 |
| 6 | Muse Spark 1.3 | Meta | 64.9 | Max effort, Loop harness | Mercor APEX-Accounting leaderboard Official board, 1 Nov 2026 |
| 7 | GPT-6 Astra | OpenAI | 64.7 | Max effort, Loop harness | Mercor APEX-Accounting leaderboard Official board, 1 Nov 2026 |
| 8 | Sonnet 5.5 | Anthropic | 64.5 | Max effort, Loop harness | Mercor APEX-Accounting leaderboard Official board, 1 Nov 2026 |
| 9 | GPT-6.1 Sol | OpenAI | 64.3 | Max effort, Loop harness | Mercor APEX-Accounting leaderboard Official board, 1 Nov 2026 |
| 9 | Opus 5 | Anthropic | 64.3 | Max effort, Loop harness | Mercor APEX-Accounting leaderboard Official board, 1 Nov 2026 |
Excel Modeling Benchmark (EMB) v1.1
accuracy %, higher is better. Top 5 of 10 models.
Updated 29 Oct 2026
- 1Claude Fable 5.176.7
- 2Gemini 4 Carbon76.5
- 3Claude Sonnet 5.575.9
- 4Claude Opus 5.575.8
- 5Wonder Full75
| 1 | Claude Fable 5.1 | Anthropic | 76.7 | v1.1 test set | Vals.ai leaderboard Official board, 29 Oct 2026 |
|---|---|---|---|---|---|
| 2 | Gemini 4 Carbon | 76.5 | v1.1 test set | Vals.ai leaderboard Official board, 29 Oct 2026 | |
| 3 | Claude Sonnet 5.5 | Anthropic | 75.9 | v1.1 test set | Vals.ai leaderboard Official board, 29 Oct 2026 |
| 4 | Claude Opus 5.5 | Anthropic | 75.8 | v1.1 test set | Vals.ai leaderboard Official board, 29 Oct 2026 |
| 5 | Wonder Full | Not stated | 75 | v1.1 test set | Vals.ai leaderboard Official board, 29 Oct 2026 |
| 6 | Gemini 4 Argon | 74.8 | v1.1 test set | Vals.ai leaderboard Official board, 29 Oct 2026 | |
| 7 | GPT-6.1 Terra | OpenAI | 73.3 | v1.1 test set | Vals.ai leaderboard Official board, 29 Oct 2026 |
| 8 | Gemini 3.8 Flash | 72.9 | v1.1 test set | Vals.ai leaderboard Official board, 29 Oct 2026 | |
| 9 | GPT-6 Astra | OpenAI | 72.7 | v1.1 test set | Vals.ai leaderboard Official board, 29 Oct 2026 |
| 10 | Wonder | Not stated | 72.5 | v1.1 test set | Vals.ai leaderboard Official board, 29 Oct 2026 |
BlueFin arXiv v1 (2026-05-29)
criteria-weighted score % (overall), higher is better. 5 models.
Updated May 2026
- 1GPT 5.549.6
- 2Opus 4.749.2
- 3gemini-3.1-pro45.9
- 4Sonnet 4.642.9
- 5Grok 4.2031.3
| 1 | GPT 5.5 | OpenAI | 49.6 | held-out set, 120 tasks | Paper Table 2 Reported by the authors, May 2026 |
|---|---|---|---|---|---|
| 2 | Opus 4.7 | Anthropic | 49.2 | held-out set, 120 tasks | Paper Table 2 Reported by the authors, May 2026 |
| 3 | gemini-3.1-pro | 45.9 | held-out set, 120 tasks | Paper Table 2 Reported by the authors, May 2026 | |
| 4 | Sonnet 4.6 | Anthropic | 42.9 | held-out set, 120 tasks | Paper Table 2 Reported by the authors, May 2026 |
| 5 | Grok 4.20 | xAI | 31.3 | held-out set, 120 tasks | Paper Table 2 Reported by the authors, May 2026 |
Finch: Finance & Accounting Cowork Benchmark arXiv v5 (2026-04-15)
workflow pass rate % (human eval), higher is better. 2 models.
Updated Apr 2026
- 1GPT 5.1 Pro38.4
- 2claude-sonnet-4.525
| 1 | GPT 5.1 Pro | OpenAI | 38.4 | product agent, 172 workflows | Paper Table 5 Reported by the authors, Apr 2026 |
|---|---|---|---|---|---|
| 2 | claude-sonnet-4.5 | Anthropic | 25 | product agent, 172 workflows | Paper Table 5 Reported by the authors, Apr 2026 |