CoCoBench: leaderboard

Metric: Task success rate (%; share of episodes whose goal predicates all hold at the end; 897 oracle-validated executable household multi-agent task instances with two to four agents, centralized planner with image observations, task-specific step budgets). Source: arxiv.org. Saturation forecast: Around December 2026. 11 models tracked.

Top models

#ModelScore
1GPT-5.6 Sol84.8
2Claude Opus 4.878.5
3Qwen 3.6 Plus69.1
4Claude Haiku 4.563.1
5Kimi K2.657.6
6Qwen 3.5 9B36.7
7GPT-5.4 Mini33.6
8Qwen 3 VL 8B Instruct24.6

Interactive version: theaggregate.ai/benchmark?slug=cocobench · How It Works · Data refreshed daily, snapshot 2026-09-26.