Office Comprehension Bench - File Fidelity Word: leaderboard
Perceiving structural and visual content in native .docx files — tables, figures and formatting — answered from the file itself; the widest model spread of the three applications.
Metric: File Fidelity Accuracy (%). Source: arxiv.org. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 | 91.5 | #49 |
| 2 | Median Human | 78.6 | |
| 3 | GPT-5.5 | 68.7 | #31 |
| 4 | Gemini 3.1 Pro (Preview) | 55.3 | #58 |
Interactive version: theaggregate.ai/benchmark?slug=office-comprehension-bench-file-fidelity-word · How It Works · Data refreshed daily, snapshot 2026-10-20.