Excel Modeling Benchmark Operating
Professional Work · 2026-07-01
This row reports the category score for operating-model tasks in Excel Modeling Benchmark.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5 | 73.5 |
| Claude Fable 5 | 72.8 |
| GPT-5.6 Luna | 72.7 |
| GPT-5.6 Sol | 67.3 |
| Opus 4.8 | 66.7 |
| Sonnet 5 | 65.5 |
| GPT-5.5 | 64.5 |
| Gemini 3.6 Flash | 63.0 |
| GLM-5.2 | 62.9 |
| Grok 4.5 | 61.9 |
| GPT-5.6 Terra | 60.4 |
| Kimi K3 | 59.4 |
| Gemini 3.5 Flash | 58.7 |
| Gemini 3.1 Pro Preview | 58.1 |
| Sonnet 4.6 | 58.1 |
| Qwen3.7-Plus | 58.0 |
| Muse Spark 1.1 | 57.6 |
| Kimi K2.6 | 56.9 |
| MiMo-V2.5-Pro | 55.2 |
| Qwen3.7-Max | 52.1 |
| GPT-5.4 Mini | 50.8 |
| MiniMax M3 | 48.9 |
| DeepSeek-V4-Pro | 47.2 |
| Inkling | 41.6 |
| GPT-5.4 Nano | 36.7 |
| Inkling-Small | 36.4 |
| Gemini 3.5 Flash-Lite | 32.7 |
| Grok 4.3 | 16.3 |
| Gemini 3.1 Flash-Lite Preview | 5.4 |