Wikibench

OfficeQA Pro

reasoning

Multilingual: no

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.

Methodology

Imported from llm-stats public benchmark metadata. Modality: text. Max score: 1. Categories: agents, general, reasoning. Language: en. Verified by llm-stats: no.

Leaderboard

Showing 3 of 3 results

  1. Claude Opus 4.8 self-reported llm-stats
    66.2%
  2. GPT-5.5 self-reported llm-stats
    54.1%
  3. MiniMax M3 self-reported llm-stats
    45.1%