Excel Modeling Benchmark Misc
Professional Work · 2026-07-01
This row reports the category score for miscellaneous financial-modeling tasks in Excel Modeling Benchmark.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Fable 5 | 71.4 |
| GPT-5.6 Luna | 70.1 |
| GPT-5.6 Sol | 68.7 |
| Claude Opus 5 | 66.9 |
| GPT-5.5 | 64.6 |
| Gemini 3.6 Flash | 64.4 |
| Kimi K3 | 63.8 |
| Gemini 3.5 Flash | 62.2 |
| GPT-5.6 Terra | 61.8 |
| Sonnet 4.6 | 61.7 |
| GLM-5.2 | 60.8 |
| Opus 4.8 | 59.4 |
| Qwen3.7-Max | 57.7 |
| DeepSeek-V4-Pro | 57.6 |
| Kimi K2.6 | 54.6 |
| Sonnet 5 | 52.3 |
| MiMo-V2.5-Pro | 50.6 |
| MiniMax M3 | 50.0 |
| Grok 4.5 | 49.2 |
| Gemini 3.5 Flash-Lite | 48.8 |
| Muse Spark 1.1 | 48.0 |
| Inkling | 46.6 |
| GPT-5.4 Mini | 46.3 |
| Gemini 3.1 Pro Preview | 43.4 |
| GPT-5.4 Nano | 43.2 |
| Qwen3.7-Plus | 42.3 |
| Inkling-Small | 36.3 |
| Grok 4.3 | 18.5 |
| Gemini 3.1 Flash-Lite Preview | 8.7 |