Causal-Plan-Bench - Effects: leaderboard

Metric: Effects score: mean four-choice accuracy (%) after answer normalization over the affordance visual semantics, spatial postcondition and affordance postcondition tasks; 100 held-out expert-reviewed items per task drawn from egocentric and robot video (EPIC-KITCHENS, Ego4D, HoloAssist and others), the same instruction, visual evidence and answer format for every model, temperature 0.2, mean of three regenerated runs; higher is better. Source: arxiv.org. Saturation forecast: Around 2036. 14 models tracked.

Top models

#ModelScore
1GPT-639.67
2Seed 2.1 Pro37.33
3Gemini 3.1 Pro (Preview)36
4GPT-5.6 Sol36
5Kimi K2.533.33
6Qwen 3 VL 8B Instruct33
7InternVL3.5-8B31.67
8GPT-4o30.33

Interactive version: theaggregate.ai/benchmark?slug=causal-plan-bench-effects · How It Works · Data refreshed daily, snapshot 2026-09-29.