Excel Modeling Benchmark leaderboard
As of 6 Nov 2026, the top score on Excel Modeling Benchmark is 76.7% by Claude Fable 5.1 (Anthropic), ahead of Claude Opus 5.5 at 75.9%, across 75 model configurations with a published result (Vals AI, result dated 20 Oct 2026). The model must produce a working Excel financial model, with formulas and structure, from an analyst-style brief. Excel Modeling Benchmark is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 6 Nov 2026)
- Claude Fable 5.1 holds the top Excel Modeling Benchmark score at 76.7% (Vals AI, 20 Oct 2026).
- Claude Opus 5.5 is second on Excel Modeling Benchmark at 75.9% (Vals AI, 20 Oct 2026).
- Excel Modeling Benchmark: Building financial models in Excel from a brief. Run by Vals AI (Vals AI).
- 75 model configurations have a published Excel Modeling Benchmark result on BenchLeader (BenchLeader, 6 Nov 2026).
- Excel Modeling Benchmark does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
Building financial models in Excel from a brief. Run by Vals AI.
- Published by
- Vals AI
- Category
- Knowledge
- Index weight
- Reference only
- Models
- 75
- Data as of
- 6 Nov 2026
Vals AI (vals.ai).
What the test looks like
The model must produce a working Excel financial model, with formulas and structure, from an analyst-style brief.
How it is scored
Rubric score across models built, run by Vals AI.
What to keep in mind
Tool access and file handling matter as much as finance knowledge.
- 1Claude Fable 5.176.7%
- 2Claude Opus 5.575.9%
- 3Claude Sonnet 5.575.7%
- 4Gemini 4 Argon (high)75.2%
- 5Claude Fable 573.7%
- 6Claude Opus 573.6%
- 7GPT-5.6 Sol (max)72.3%
- 8Gemini 3.8 Flash (high)72.2%
- 9Qwen3.8 Max Internal72.0%
- 10Muse Wonder Full (max)71.9%
- 11GPT-6 Astra (max)71.7%
- 12GPT-6 Sol (max)71.5%
- 13Gemini 3.7 Flash (high)71.3%
- 14GPT-6.1 Sol (max)70.8%
- 15Claude Opus 4.869.4%
75 of 75
| # | |||||
|---|---|---|---|---|---|
| 1 |  Claude Fable 5.1 | 76.7% | 66.4 | $15.93 | 2026-10-20 |
| 2 | Claude Opus 5.5 | 75.9% | 63.4 | $11.02 | 2026-10-20 |
| 3 | Claude Sonnet 5.5 | 75.7% | – | $8.79 | 2026-10-20 |
| 4 |  Gemini 4 Argonhigh | 75.2% | 66.0 | $11.80 | 2026-10-20 |
| 5 | Claude Fable 5 | 73.7% | 66.5 | $12.35 | 2026-10-20 |
| 6 | Claude Opus 5 | 73.6% | 65.2 | $8.02 | 2026-10-20 |
| 7 |  GPT-5.6 Solmax | 72.3% | 65.4 | $6.01 | 2026-10-20 |
| 8 | Gemini 3.8 Flashhigh | 72.2% | 62.0 | $8.24 | 2026-10-20 |
| 9 |  Qwen3.8 Max Internal | 72.0% | – | $4.43 | 2026-10-20 |
| 10 |  Muse Wonder Fullmax | 71.9% | 62.9 | $4.58 | 2026-10-20 |
| 11 | GPT-6 Astramax | 71.7% | 67.8 | $5.83 | 2026-10-20 |
| 12 | GPT-6 Solmax | 71.5% | 63.3 | $1.28 | 2026-10-20 |
| 13 | Gemini 3.7 Flashhigh | 71.3% | 62.2 | $5.69 | 2026-10-20 |
| 14 | GPT-6.1 Solmax | 70.8% | 67.3 | $1.14 | 2026-10-20 |
| 15 | Claude Opus 4.8 | 69.4% | 60.0 | $12.06 | 2026-10-20 |
| 16 | GPT-6 Lunamax | 68.5% | 56.7 | $0.142 | 2026-10-20 |
| 17 | Muse Spark 1.3max | 67.4% | 62.0 | $2.55 | 2026-10-20 |
| 18 | GPT-5.6 Lunamax | 67.1% | 56.3 | $0.367 | 2026-10-20 |
| 19 | Muse Wondermax | 67.1% | 62.3 | $5.60 | 2026-10-20 |
| 20 |  Grok 4.7xhigh | 67.0% | 58.3 | $6.48 | 2026-10-20 |
| 21 |  Kimi K3maxopen ↗ | 66.7% | 62.1 | $3.07 | 2026-10-20 |
| 22 | Claude Sonnet 5 | 66.3% | 55.9 | $10.29 | 2026-10-20 |
| 23 | GPT-5.6 Terramax | 66.2% | 60.7 | $2.19 | 2026-10-20 |
| 24 |  MiMo-V2.6-Flashopen ↗ | 65.5% | 55.0 | $0.109 | 2026-10-20 |
| 25 | Gemini 3.6 Flashhigh | 65.4% | 61.2 | $3.89 | 2026-10-20 |
| 26 | GPT-5.5xhigh | 64.5% | 64.3 | $3.27 | 2026-10-20 |
| 27 | Claude Opus 4.7 | 63.7% | 63.1 | $12.76 | 2026-10-20 |
| 28 | Gemini 3.5 Flashhigh | 63.5% | 62.5 | $5.17 | 2026-10-20 |
| 29 | Ember-1 | 63.4% | – | $2.46 | 2026-10-20 |
| 30 | MiMo-V2.6-Proopen ↗ | 62.9% | 57.3 | $0.329 | 2026-10-20 |
| 31 | Grok 4.6high | 62.7% | 60.5 | $3.06 | 2026-10-20 |
| 32 | Muse Spark 1.3xhigh | 62.7% | 58.5 | $3.55 | 2026-10-20 |
| 33 |  DeepSeek V4.1 Flash 1017maxopen ↗ | 61.8% | 59.2 | $0.355 | 2026-10-20 |
| 34 |  GLM 5.2open ↗ | 61.5% | 51.5 | $4.47 | 2026-10-20 |
| 35 | Claude Sonnet 4.6 | 60.1% | 58.2 | $7.23 | 2026-10-20 |
| 36 | Qwen3.8 Maxmax | 60.1% | 58.8 | $2.66 | 2026-10-20 |
| 37 | Qwen3.8 27Bxhighopen ↗ | 59.7% | 52.5 | $4.56 | 2026-10-20 |
| 38 | Claude Haiku 5.5 | 59.0% | – | $1.15 | 2026-10-20 |
| 39 |  Hy4open ↗ | 58.1% | 55.7 | $0.787 | 2026-10-20 |
| 40 | Kimi K2.6open ↗ | 57.9% | 55.2 | $2.20 | 2026-10-20 |
| 41 | DeepSeek V4.1 Flashhighopen ↗ | 57.2% | 55.7 | $0.234 | 2026-10-20 |
| 42 |  Step 5 Previewopen | 57.1% | 54.9 | $1.33 | 2026-10-20 |
| 43 | Qwen3.7max | 57.0% | 60.3 | $2.86 | 2026-10-20 |
| 44 | Muse Spark 1.2xhigh | 57.0% | 61.7 | $2.31 | 2026-10-20 |
| 45 | DeepSeek V4 Flashhighopen ↗ | 57.0% | 57.3 | $0.384 | 2026-10-20 |
| 46 | Muse Spark 1.1xhigh | 56.4% | 59.5 | $1.11 | 2026-10-20 |
| 47 | GLM 5.3maxopen ↗ | 56.3% | 59.0 | $3.79 | 2026-10-20 |
| 48 |  Mistral Large 4highopen ↗ | 56.2% | 49.5 | $4.84 | 2026-10-20 |
| 49 | GLM 5.3 Flashmaxopen ↗ | 55.9% | 54.2 | $0.713 | 2026-10-20 |
| 50 | MiMo-V2.5-Proopen ↗ | 55.2% | 54.9 | $0.221 | 2026-10-20 |
| 51 | Muse Spark Openxhighopen ↗ | 53.7% | 54.0 | $5.35 | 2026-10-20 |
| 52 | Grok 4.5high | 52.9% | 56.9 | $1.55 | 2026-10-20 |
| 53 | DeepSeek V4 Promaxopen ↗ | 52.8% | 57.0 | $1.08 | 2026-10-20 |
| 54 | Gemini 3.1 Prohigh | 52.6% | 60.1 | $3.77 | 2026-10-20 |
| 55 | Qwen3.7 Plus | 49.4% | 57.7 | $0.598 | 2026-10-20 |
| 56 |  MiniMax M3open ↗ | 47.8% | 56.4 | $2.09 | 2026-10-20 |
| 57 | GPT-5.4 Minixhigh | 45.4% | 56.0 | $2.15 | 2026-10-20 |
| 58 | GPT-5.4 Nanohigh | 44.8% | 48.3 | $0.573 | 2026-10-20 |
| 59 | Gemini 3.5 Flash Litehigh | 43.2% | 47.9 | $1.23 | 2026-10-20 |
| 60 | Inklingopen ↗ | 38.4% | 47.5 | $1.67 | 2026-10-20 |
More knowledge benchmarks · All LLM benchmarks
Related benchmarks
Top head-to-heads on Excel Modeling Benchmark
Cite as: BenchLeader, “Excel Modeling Benchmark leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/vals_emb.
Excel Modeling Benchmark: questions
- What does Excel Modeling Benchmark measure?
- The model must produce a working Excel financial model, with formulas and structure, from an analyst-style brief. Scores are reported in percent of tasks solved; higher is better.
- What is the top score on Excel Modeling Benchmark?
- As of 6 Nov 2026, the highest score on Excel Modeling Benchmark is 76.7% by Claude Fable 5.1 (result dated 20 Oct 2026), ahead of Claude Opus 5.5 at 75.9%. Scores come from Vals AI and are re-read daily.
- Which AI model leads Excel Modeling Benchmark?
- Claude Fable 5.1 leads Excel Modeling Benchmark with 76.7% as of 6 Nov 2026 (result dated 20 Oct 2026), ahead of Claude Opus 5.5 at 75.9%.
- How many models have Excel Modeling Benchmark results?
- 75 model configurations have a Excel Modeling Benchmark result on BenchLeader, all taken from Vals AI.
- Who runs Excel Modeling Benchmark and how often is it updated?
- Excel Modeling Benchmark is published by Vals AI. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does Excel Modeling Benchmark count toward the BenchLeader Index?
- No. Excel Modeling Benchmark is shown for reference but left out of the composite index.