Mind2Act World (Agents) - Overall Success Rate: leaderboard

Overall episode success rate (%) for zero-shot multimodal API agents under one fixed motion-tool harness on Mind2Act World (arXiv:2611.26444v1, Table 4; 900 episodes per agent across 30 tasks, 3 difficulty levels, and Clean/Random settings).

Metric: Overall episode success rate (%). Source: arxiv.org. Saturation forecast: Around 2029. 3 models tracked. Updated 21 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-6 AstraMedium34.6#10
2DeepSeek V4.1 FlashMedium6.6#67
3Gemini 4 ArgonMedium4.7#19

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=mind2act-world-agents-overall-success-rate · How It Works · Data refreshed daily, snapshot 2026-11-21.