Coding Rankings

Text Capability Rankings

Evaluate models on text comprehension and writing.

Based on a shared set of Chinese tasksMethodology
Scoring dimension
About the scores

The same Chinese tasks and scoring criteria are used to compare comprehension and finished drafts.

Methodology
RankHigh to low
Ranking order

Loading rankings…

Evaluation methodology

The same Chinese tasks and scoring criteria are used to compare comprehension and finished drafts.