ChartAct - Dynamic Chart: leaderboard

ChartAct (2026) Dynamic Chart environment: a multimodal agent interacts through screenshots with a real interactive web chart shown alone on a clean page to answer questions whose evidence appears only after hovering, clicking, zooming or dragging. Success rate in percent; higher is better.

Metric: Success rate (%) in the Dynamic Chart environment (the chart alone on a clean page); the model acts in an OSWorld-style desktop on screenshots only (pyautogui actions, at most 15 steps, temperature 0.3) to hover, click, zoom or drag 673 real interactive web charts and answers 1,440 human-reviewed questions, an LLM judge with majority voting deciding correctness; Claude Opus 4.7, GPT-5.5 and Gemini 3.1 Pro are scored on a representative 300-case subset per environment; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Opus 4.785.3#85
2Kimi K2.5Non-reasoning75.3#266
3GPT-5.565#54
4Gemini 3.1 Pro (Preview)Low61#99
5Doubao-Seed-2-0-Pro-260215High55.6
6Qwen 3.6 PlusThinking54.7#238
7Qwen 3.5 122B A10B51.9#346
8Qwen 3.5 35B A3B45.7#492
9GLM-4.6V38.7#710
10Qwen 3 VL PlusThinking31.4#512
11Gemma 4 31B (IT)31.3#350

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=chartact-dynamic-chart · How It Works · Data refreshed daily, snapshot 2026-11-21.