reasoning benchmark
OfficeQA Pro
OfficeQA Pro evaluates AI models on professional knowledge-work questions and tasks drawn from real office workflows, including document analysis, spreadsheet reasoning, and information synthesis across business domains.
Updated Aug 7, 2026
Published models7
Registry coverage7
MetricScore
EvidenceB
OfficeQA Pro leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | BY | 72.2% | 100.0% | 7 | C | |
| 2 | BY | 71.1% | 83.3% | 7 | C | |
| 3 | AN | 66.2% | 66.7% | 7 | C | |
| 4 | MA | 63.3% | 50.0% | 7 | C | |
| 5 | AN | 59.4% | 33.3% | 7 | C | |
| 6 | OP | 54.1% | 16.7% | 7 | C | |
| 7 | MI | 45.1% | 0.0% | 7 | C |
Score distribution
Top published rows on the benchmark's original scale.
OfficeQA Pro highlights
The top published results on this benchmark's own scale.
What is OfficeQA Pro?
Definition and scoring fields from the benchmark registry.
OfficeQA Pro evaluates AI models on professional knowledge-work questions and tasks drawn from real office workflows, including document analysis, spreadsheet reasoning, and information synthesis across business domains.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- OfficeQA Pro
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- officeqa-pro|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about OfficeQA Pro.
Which model scores highest on OfficeQA Pro?
Seed 2.1 Pro is currently ranked first with 72.2%.
What does OfficeQA Pro measure?
OfficeQA Pro evaluates AI models on professional knowledge-work questions and tasks drawn from real office workflows, including document analysis, spreadsheet reasoning, and information synthesis across business domains.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
7 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.