# OfficeQA Pro (extracted text) — AI model rankings

answering the 133 Treasury document questions with extracted text and code execution available

9 tracked models have a published OfficeQA Pro (extracted text) score. Higher is better. Scores come from published lab reports and benchmark sources; recorded sources appear beside the scores.

## Ranking

| Rank | Model | Developer | Score | Source | Released |
| --- | --- | --- | --- | --- | --- |
| 1 | Claude Fable 5.1 | Anthropic | 69% | Lab | Sep 1 2026 |
| 2 | Claude Opus 5.5 | Anthropic | 67.7% | Lab | Sep 22 2026 |
| 3 | Claude Mythos 5 | Anthropic | 67.1% | Lab | Jun 9 2026 |
| 4 | Claude Opus 5 | Anthropic | 66.9% | Lab | Jul 24 2026 |
| 5 | Claude Opus 4.8 | Anthropic | 66.2% | Lab | May 28 2026 |
| 6 | Claude Sonnet 5.5 | Anthropic | 65.6% | Lab | Sep 28 2026 |
| 7 | Claude Opus 4.7 | Anthropic | 65% | Lab | Apr 16 2026 |
| 8 | Claude Haiku 5.5 | Anthropic | 60.3% | Lab | Oct 7 2026 |
| 9 | Claude Haiku 4.5 | Anthropic | 47.1% | Lab | Oct 15 2025 |


---

Canonical page: https://aireleasetracker.com/benchmark/officeqa-pro-extracted-text
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.