SpatialWorld - Physical: leaderboard

Metric: Task success rate (%) on all 655 physical-simulator tasks (task-weighted over the five scenario categories); task success rate (%) judged by terminal-state verifiers, vision-only egocentric observation and a unified text action interface, temperature 1.0, 30-turn history, step budget twice the human golden action count plus 10; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 15 models tracked.

Top models

#ModelScore
1GPT-514.4
2Qwen 3.5 397B A17B12.2
3Gemini 3.1 Pro (Preview)9.2
4Kimi K2.59.2
5Gemini 3 Flash7.2
6Qwen 3 VL 235B A22B Instruct6.9
7Gemini 2.5 Pro6.7
8GPT-5.46.6
9Qwen 3 VL 235B A22B (Thinking)6.1
10Seed 2.0 Lite5.8
11GLM-4.5V4
12Qwen 2.5 VL 72B Instruct3.4
13GLM-4.6V2.7

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