# Best AI for Financial spreadsheets and accounting (November 2026)

The leader on each benchmark in this use case, never combined. Data refreshed 2 Nov 2026.

## Picks

- APEX-Accounting: Gemini 4 Argon: 71.5 Mean Score (% criteria passed) (Mercor APEX-Accounting leaderboard, 1 Nov 2026), https://benchmarks.finance/leaderboards/apex-accounting/
- Excel Modeling Benchmark (EMB): Claude Fable 5.1: 76.7 accuracy % (v1.1) (Vals.ai leaderboard, 29 Oct 2026), https://benchmarks.finance/leaderboards/vals-emb/
- BlueFin: GPT-5.5: 49.6 criteria-weighted score % (overall) (Paper Table 2, May 2026), https://benchmarks.finance/leaderboards/bluefin/
- Ramp Accounting Bench: Gemini 4 Argon: 61.1 mean reward % (criteria, mean of 3 attempts) (Ramp Labs Accounting Bench, 14 Oct 2026), https://benchmarks.finance/leaderboards/ramp-accounting-bench/
- Finch: Finance & Accounting Cowork Benchmark: GPT 5.1 Pro: 38.4 workflow pass rate % (human eval) (Paper Table 5, Apr 2026), https://benchmarks.finance/leaderboards/finch/

## APEX-Accounting: Mean Score (% criteria passed) (higher is better, 10 models)

| Rank | Model | Provider | Score | Metric | Setting | Reported by | Source | URL | Date |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | Gemini 4 Argon | Google | 71.5 | Mean Score (% criteria passed) | High effort, Loop harness | Official board | Mercor APEX-Accounting leaderboard | https://www.mercor.com/apex/apex-accounting-leaderboard/ | 1 Nov 2026 (checked 1 Nov 2026) |
| 2 | Gemini 4 Carbon | Google | 70.7 | Mean Score (% criteria passed) | High effort, Loop harness | Official board | Mercor APEX-Accounting leaderboard | https://www.mercor.com/apex/apex-accounting-leaderboard/ | 1 Nov 2026 (checked 1 Nov 2026) |
| 3 | Muse Wonder 1.0 | Meta | 67.1 | Mean Score (% criteria passed) | Max effort, Loop harness | Official board | Mercor APEX-Accounting leaderboard | https://www.mercor.com/apex/apex-accounting-leaderboard/ | 1 Nov 2026 (checked 1 Nov 2026) |
| 4 | Fable 5.1 | Anthropic | 66.7 | Mean Score (% criteria passed) | Max effort, Loop harness | Official board | Mercor APEX-Accounting leaderboard | https://www.mercor.com/apex/apex-accounting-leaderboard/ | 1 Nov 2026 (checked 1 Nov 2026) |
| 5 | Opus 5.5 | Anthropic | 65.5 | Mean Score (% criteria passed) | Max effort, Loop harness | Official board | Mercor APEX-Accounting leaderboard | https://www.mercor.com/apex/apex-accounting-leaderboard/ | 1 Nov 2026 (checked 1 Nov 2026) |
| 6 | Muse Spark 1.3 | Meta | 64.9 | Mean Score (% criteria passed) | Max effort, Loop harness | Official board | Mercor APEX-Accounting leaderboard | https://www.mercor.com/apex/apex-accounting-leaderboard/ | 1 Nov 2026 (checked 1 Nov 2026) |
| 7 | GPT-6 Astra | OpenAI | 64.7 | Mean Score (% criteria passed) | Max effort, Loop harness | Official board | Mercor APEX-Accounting leaderboard | https://www.mercor.com/apex/apex-accounting-leaderboard/ | 1 Nov 2026 (checked 1 Nov 2026) |
| 8 | Sonnet 5.5 | Anthropic | 64.5 | Mean Score (% criteria passed) | Max effort, Loop harness | Official board | Mercor APEX-Accounting leaderboard | https://www.mercor.com/apex/apex-accounting-leaderboard/ | 1 Nov 2026 (checked 1 Nov 2026) |
| 9 | GPT-6.1 Sol | OpenAI | 64.3 | Mean Score (% criteria passed) | Max effort, Loop harness | Official board | Mercor APEX-Accounting leaderboard | https://www.mercor.com/apex/apex-accounting-leaderboard/ | 1 Nov 2026 (checked 1 Nov 2026) |
| 9 | Opus 5 | Anthropic | 64.3 | Mean Score (% criteria passed) | Max effort, Loop harness | Official board | Mercor APEX-Accounting leaderboard | https://www.mercor.com/apex/apex-accounting-leaderboard/ | 1 Nov 2026 (checked 1 Nov 2026) |

## Excel Modeling Benchmark (EMB): accuracy % (v1.1) (higher is better, 10 models)

| Rank | Model | Provider | Score | Metric | Setting | Reported by | Source | URL | Date |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | Claude Fable 5.1 | Anthropic | 76.7 | accuracy % (v1.1) | v1.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/emb-v1-1 | 29 Oct 2026 (checked 1 Nov 2026) |
| 2 | Gemini 4 Carbon | Google | 76.5 | accuracy % (v1.1) | v1.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/emb-v1-1 | 29 Oct 2026 (checked 1 Nov 2026) |
| 3 | Claude Sonnet 5.5 | Anthropic | 75.9 | accuracy % (v1.1) | v1.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/emb-v1-1 | 29 Oct 2026 (checked 1 Nov 2026) |
| 4 | Claude Opus 5.5 | Anthropic | 75.8 | accuracy % (v1.1) | v1.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/emb-v1-1 | 29 Oct 2026 (checked 1 Nov 2026) |
| 5 | Wonder Full | Not stated | 75 | accuracy % (v1.1) | v1.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/emb-v1-1 | 29 Oct 2026 (checked 1 Nov 2026) |
| 6 | Gemini 4 Argon | Google | 74.8 | accuracy % (v1.1) | v1.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/emb-v1-1 | 29 Oct 2026 (checked 1 Nov 2026) |
| 7 | GPT-6.1 Terra | OpenAI | 73.3 | accuracy % (v1.1) | v1.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/emb-v1-1 | 29 Oct 2026 (checked 1 Nov 2026) |
| 8 | Gemini 3.8 Flash | Google | 72.9 | accuracy % (v1.1) | v1.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/emb-v1-1 | 29 Oct 2026 (checked 1 Nov 2026) |
| 9 | GPT-6 Astra | OpenAI | 72.7 | accuracy % (v1.1) | v1.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/emb-v1-1 | 29 Oct 2026 (checked 1 Nov 2026) |
| 10 | Wonder | Not stated | 72.5 | accuracy % (v1.1) | v1.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/emb-v1-1 | 29 Oct 2026 (checked 1 Nov 2026) |

## BlueFin: criteria-weighted score % (overall) (higher is better, 5 models)

| Rank | Model | Provider | Score | Metric | Setting | Reported by | Source | URL | Date |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | GPT-5.5 | OpenAI | 49.6 | criteria-weighted score % (overall) | held-out set, 120 tasks | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2605.30907v1 | May 2026 (checked 1 Nov 2026) |
| 2 | Opus 4.7 | Anthropic | 49.2 | criteria-weighted score % (overall) | held-out set, 120 tasks | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2605.30907v1 | May 2026 (checked 1 Nov 2026) |
| 3 | Gemini 3.1 Pro | Google | 45.9 | criteria-weighted score % (overall) | held-out set, 120 tasks | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2605.30907v1 | May 2026 (checked 1 Nov 2026) |
| 4 | Sonnet 4.6 | Anthropic | 42.9 | criteria-weighted score % (overall) | held-out set, 120 tasks | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2605.30907v1 | May 2026 (checked 1 Nov 2026) |
| 5 | Grok 4.20 | xAI | 31.3 | criteria-weighted score % (overall) | held-out set, 120 tasks | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2605.30907v1 | May 2026 (checked 1 Nov 2026) |

## Ramp Accounting Bench: mean reward % (criteria, mean of 3 attempts) (higher is better, 5 models)

| Rank | Model | Provider | Score | Metric | Setting | Reported by | Source | URL | Date |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | Gemini 4 Argon | Google | 61.1 | mean reward % (criteria, mean of 3 attempts) | 137 tasks, ReAct harness, GPT 5.6 Luna judge | Official board | Ramp Labs Accounting Bench | https://labs.ramp.com/accountingbench | 14 Oct 2026 (checked 2 Nov 2026) |
| 2 | Claude Opus 5.5 | Anthropic | 50.5 | mean reward % (criteria, mean of 3 attempts) | 137 tasks, ReAct harness, GPT 5.6 Luna judge | Official board | Ramp Labs Accounting Bench | https://labs.ramp.com/accountingbench | 21 Sept 2026 (checked 2 Nov 2026) |
| 3 | Claude Fable 5.1 | Anthropic | 49.7 | mean reward % (criteria, mean of 3 attempts) | 137 tasks, ReAct harness, GPT 5.6 Luna judge | Official board | Ramp Labs Accounting Bench | https://labs.ramp.com/accountingbench | 2 Sept 2026 (checked 2 Nov 2026) |
| 4 | GPT-6 Astra | OpenAI | 48.7 | mean reward % (criteria, mean of 3 attempts) | 137 tasks, ReAct harness, GPT 5.6 Luna judge | Official board | Ramp Labs Accounting Bench | https://labs.ramp.com/accountingbench | 5 Sept 2026 (checked 2 Nov 2026) |
| 4 | Muse Wonder | Meta | 48.7 | mean reward % (criteria, mean of 3 attempts) | 137 tasks, ReAct harness, GPT 5.6 Luna judge | Official board | Ramp Labs Accounting Bench | https://labs.ramp.com/accountingbench | 31 Oct 2026 (checked 2 Nov 2026) |

## Finch: Finance & Accounting Cowork Benchmark: workflow pass rate % (human eval) (higher is better, 2 models)

| Rank | Model | Provider | Score | Metric | Setting | Reported by | Source | URL | Date |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | GPT 5.1 Pro | OpenAI | 38.4 | workflow pass rate % (human eval) | product agent, 172 workflows | Reported by the authors | Paper Table 5 | https://arxiv.org/html/2512.13168v5 | Apr 2026 (checked 1 Nov 2026) |
| 2 | Claude Sonnet 4.5 | Anthropic | 25 | workflow pass rate % (human eval) | product agent, 172 workflows | Reported by the authors | Paper Table 5 | https://arxiv.org/html/2512.13168v5 | Apr 2026 (checked 1 Nov 2026) |

## Also in this use case (no dated scores yet)

- AA-AnalystAgent (Artificial Analysis): https://benchmarks.finance/benchmarks/aa-analyst-agent/
- AccountingBench (Penrose): https://benchmarks.finance/benchmarks/penrose-accountingbench/

Source: https://benchmarks.finance/best/spreadsheets-and-accounting/

Created and maintained by Moneda.