MECoBench - Parallel (2 Agents): leaderboard

Metric: Success rate (%) in the parallel setting (every agent may enter every room; 60 total steps shared by the team) with two agents, on 96 VirtualHome household rearrangement tasks (2 to 7 subgoals each) in which multimodal agents act from four-view first-person observations; agents exchange one broadcast message each before acting; printed 0-1 and x100 here; higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 15 models tracked.

Top models

#ModelScore
1Qwen 3.5 9B93.8
2Qwen 3.5 27B93.8
3GPT-5.489.6
4Gemma 4 31B (IT)83.3
5Gemma 4 26B A4B (IT)81.2
6Qwen 3 VL 32B Instruct80.2
7Gemini 3.1 Pro (Preview)77.1
8GLM-4.6V71.9
9Qwen 3 VL 235B A22B Instruct66.7
10Qwen 3 VL 8B Instruct45.8
11Llama 4 Scout Instruct25

Interactive version: theaggregate.ai/benchmark?slug=mecobench-parallel-2-agents · How It Works · Data refreshed daily, snapshot 2026-09-29.