HG-Bench - Unified Score: leaderboard

Metric: Unified score (0-100): per-sample weighted combination of question-level and step-level page F1 averaged over all 500 samples, unparseable outputs scoring 0; 500 human-annotated multi-page handwritten K-12 homework samples; the model outputs page-aware question-level answer boxes and ordered step boxes as JSON with normalized 0-1000 coordinates, one shared prompt, default decoding, one format-reminder retry; higher is better. Source: arxiv.org. Saturation forecast: Around 2030. 9 models tracked.

Top models

#ModelScore
1Gemini 3 Pro (Preview)42.33
2GLM-5V Turbo40.1
3Qwen 3.5 397B A17B32.73
4Doubao-Seed-2.0-Pro-26021521.22
5Kimi K2.520.18
6Claude Sonnet 4.68.76
7GPT-5.48.12

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