FireWorldBench (Rendered Images): leaderboard

Metric: Completion accuracy (%; equal-weight mean of the five physical-capability dimensions P1-P5 on the 8,360-item controlled-simulation subset; image-observation track (physics-rendered field visualizations and 3D scene renderings); each cell is the mean of the matched choice-format score (Jaccard overlap of selected and gold option sets) and open-report score (exact agreement of required structured slots); deterministic evaluator, no LLM judge; failed or unparseable responses score 0). Source: arxiv.org. Saturation forecast: Around 2028. 10 models tracked.

Top models

#ModelScore
1Gemini 2.5 Flash44.83
2GPT-5.6 Luna42.99
3Qwen 3 VL 32B Instruct42.55
4Claude Haiku 4.539.51
5Qwen 3.5 Plus (2026-02-15)39.4
6InternVL3-78B38.72
7InternVL3-38B34.81
8Qwen 3 VL 235B A22B Instruct34.02
9Qwen 3 VL 8B Instruct29.41
10InternVL3-8B27.77

Interactive version: theaggregate.ai/benchmark?slug=fireworldbench-rendered-images · How It Works · Data refreshed daily, snapshot 2026-09-26.