Excel Modeling Benchmark DCF
Professional Work · 2026-07-01
This row reports the category score for discounted-cash-flow tasks in Excel Modeling Benchmark.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5.5 | 70.1 |
| Claude Sonnet 5.5 | 70.1 |
| Claude Opus 5 | 70.0 |
| Claude Fable 5.1 | 69.9 |
| Claude Fable 5 | 68.9 |
| Muse Wonder Full | 66.6 |
| Opus 4.8 | 66.3 |
| Gemini 3.7 Flash | 66.1 |
| Gemini 4 Argon | 65.7 |
| GPT-6.1 Sol | 65.3 |
| GPT-6 Astra | 64.5 |
| Qwen3.8 Max (0902) | 64.3 |
| Gemini 3.8 Flash | 63.7 |
| GPT-6 Sol | 63.1 |
| GPT-6 Luna | 62.8 |
| GPT-5.6 Sol | 62.8 |
| GPT-5.6 Luna | 60.5 |
| GLM-5.3 Flash | 59.9 |
| Gemini 3.6 Flash | 59.4 |
| MiMo-V2.6-Flash | 59.0 |
| Gemini 3.5 Flash | 58.4 |
| Grok 4.7 | 58.2 |
| MiMo-V2.6-Pro | 58.1 |
| Opus 4.7 | 57.1 |
| GPT-5.5 | 56.7 |
| Kimi K3 | 56.7 |
| Sonnet 5 | 56.4 |
| Muse Spark 1.3 | 55.7 |
| Muse Spark 1.2 | 55.6 |
| Claude Haiku 5.5 | 55.1 |
| GPT-5.6 Terra | 53.4 |
| DeepSeek V4.1 Flash 1017 | 53.0 |
| GLM-5.2 | 53.0 |
| MiMo-V2.5 | 52.5 |
| DeepSeek-V4-Flash-0731 | 51.3 |
| Muse Wonder | 51.0 |
| DeepSeek V4.1 Flash | 51.0 |
| Ember-1 | 50.3 |
| Qwen3.8 27B | 50.0 |
| Grok 4.6 | 49.8 |