OccuBench — leaderboard

Professional-task benchmark using simulated domain tool environments to evaluate LLM agents across occupation-specific workflows.

Metric: Completion rate (self-reported). Source: benchmarklist.com. Status: saturation imminent. 15 models tracked.

Top models

#ModelScore
1Gemini 3.1 Pro (Preview)45.3
2GPT-5.242.6
3Qwen 3.5 Plus (2026-04-20)41
4Claude Opus 4.640.1
5MiniMax-M2.739.4
6GLM-5.135.7
7GPT-4o28.8
8Llama 4 Maverick22
9Gemma 3 27B7
10DeepSeek V3.21.9
11Gemma 3 12B1.5
12Qwen 3.5 4B0.2

Interactive version: theaggregate.ai/benchmark?slug=occubench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.