基准测试

Chartography (w/tools)

Chart and figure understanding, tool-assisted.

分数历史 →

模型分数

#	模型	分数
1	Claude Opus 5.5	89%
2	Claude Opus 5	84%
3	DeepSeek V4.1-Flash 1017	83%
4	GPT 5.6-Sol	79.9%
5	DeepSeek V4.1-Flash 0910	78.9%
6	Kimi K3	68.1%