CoCoBench - Task Allocation: leaderboard
CoCoBench (2026) evaluates multimodal LLMs as centralized planners for two to four cooperating embodied agents on 897 executable household tasks. Task success rate in percent, higher is better, on the task-allocation instances, where agents must divide independent subtasks without repeated work.
Metric: Task success rate (%; task-allocation instances, dividing independent subtasks without duplicated work; 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: Not forecast. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 8B Instruct | 95 | #982 |
| 2 | Qwen 3.5 9B | 91.3 | #803 |
| 3 | Claude Haiku 4.5 | 89.4 | #568 |
| 4 | GPT-5.4 Mini | 81.7 | #389 |
| 5 | Qwen 3.6 Plus | 78.4 | #217 |
| 6 | Kimi K2.6 | 71.6 | #173 |
| 7 | GPT-5.6 Sol | 57.3 | #29 |
| 8 | Claude Opus 4.8 | 38.1 | #56 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=cocobench-task-allocation · How It Works · Data refreshed daily, snapshot 2026-11-12.