Office Comprehension Bench: leaderboard
Office Comprehension Bench (Microsoft) — Domain Q&A track: expert-level multi-step reasoning over native Word/Excel/PowerPoint files across 12 professional domains; assertion-level accuracy via LLM-judge ensemble.
Metric: Domain Q&A Accuracy (%). Source: arxiv.org. Saturation forecast: Around February 2027. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 59.35 | #31 |
| 2 | Claude Opus 4.7 | 56.83 | #49 |
| 3 | Gemini 3.1 Pro (Preview) | 45.75 | #58 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=office-comprehension-bench · How It Works · Data refreshed daily, snapshot 2026-10-20.