CLUE — leaderboard

Chinese Language Understanding Evaluation benchmark covering Chinese classification and reading-comprehension tasks with public leaderboard baselines.

Metric: Classification Score (%). Source: github.com. 12 models tracked.

Top models

#ModelScore
1RoBERTa-wwm-large72.83
2RoBERTa-large71.7
3ALBERT-xxlarge71.04
4RoBERTa-wwm-ext70.1
5ALBERT-xlarge68.92
6BERT-base68.77
7BERT-wwm-ext68.75
8XLNet-mid68.58
9ERNIE-base68.55
10ALBERT-large67.91

Interactive version: theaggregate.ai/benchmark?slug=clue · How the rankings work · Data refreshed daily, snapshot 2026-07-22.