OccuBench - Business & Enterprise: leaderboard

Metric: Completion rate (%) on the Business & Enterprise industry's tasks in the clean environment (E0); 382 tool-use task instances from 100 professional scenarios, the environment simulated by Gemini-3-Flash-Preview and each trajectory passed or failed by a rubric verifier; thinking mode on, high reasoning effort where configurable; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 15 models tracked.

Top models

#ModelScore
1GPT-5.2 (High)86
2DeepSeek V3.2 (Thinking)78
3Claude Opus 4.6 (High)78
4Claude Opus 4.5 (Thinking)76
5GLM-5 (Thinking)75
6Claude Opus 4 (Thinking)75
7Gemini 3.1 Pro (Preview)73
8Claude Sonnet 4.6 (Thinking)71
9Claude Sonnet 4.5 (Thinking)70
10Qwen 3.5 Plus (Thinking)70
11Claude Sonnet 4 (Thinking)63
12Kimi K2.5 (Thinking)62
13MiniMax-M2.760
14Qwen 3.5 Flash (Thinking)60

Interactive version: theaggregate.ai/benchmark?slug=occubench-business-enterprise · How It Works · Data refreshed daily, snapshot 2026-10-07.