ExChart-Bench - Line Charts: leaderboard

Metric: Adaptive MAPE (%; 9,494 values in 700 charts, per-value error relative to the chart's largest absolute value, capped at 100; unparsable or misshapen tables count 100). Source: arxiv.org. Saturation forecast: Estimated already saturated. 12 models tracked.

Top models

#ModelScore
1Gemini 2.5 Flash4.35
2GLM-4.5V5.56
3GPT-4.16.43
4Qwen 2.5 VL 72B Instruct6.57
5Qwen 2.5 VL 32B Instruct8.08
6GPT-4o9.53
7Qwen 2.5 VL 7B Instruct22.4

Interactive version: theaggregate.ai/benchmark?slug=exchart-bench-line-charts · How It Works · Data refreshed daily, snapshot 2026-09-25.