WritingBench

Multilingual
text
+
+
+
+
About

WritingBench is a comprehensive generative writing benchmark evaluating large language models across 6 core writing domains and 100 subdomains, including creative, persuasive, informative, and technical writing. Using a query-dependent evaluation framework with instance-specific assessment criteria and a fine-tuned critic model, this benchmark provides systematic evaluation of AI writing capabilities across diverse genres and styles.

+
+
+
+
Evaluation Stats
Total Models4
Organizations1
Verified Results0
Self-Reported4
+
+
+
+
Benchmark Details
Max Score1
Language
en
+
+
+
+
Performance Overview
Score distribution and top performers

Score Distribution

4 models
Top Score
88.3%
Average Score
86.4%
High Performers (80%+)
4

Top Organizations

#1Alibaba Cloud / Qwen Team
4 models
86.4%
+
+
+
+
Leaderboard
4 models ranked by performance on WritingBench
LicenseLinks
Jul 25, 2025
Apache 2.0
88.3%
Sep 10, 2025
Apache 2.0
87.3%
Jul 22, 2025
Apache 2.0
85.2%
Sep 10, 2025
Apache 2.0
84.6%
+
+
+
+
Resources