CCLUE — leaderboard

Classical Chinese Language Understanding Evaluation benchmark covering segmentation and punctuation, named entity recognition, classification, sentiment, and classical-modern retrieval.

Metric: Average Score (%). Source: ethan-yt.github.io. 3 models tracked.

Top models

#ModelScore
1guwenbert-base77.55
2guwenbert-base-fs77.21
3chinese-roberta-wwm-ext70.88

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