Creative Writing
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
45.2Discovery Score
12
Feb 26, 2026
8.95Metric 1 Score
9
May 12, 2026
75Human Preference Count
9
Apr 14, 2026
7.9Score (Model 4)
6
Jun 18, 2026
63.3QD-Score
6
May 12, 2026
63.3QD-Score
6
May 12, 2026
7.91GPT-4 Coherence Score
6
May 8, 2026
80.4Accuracy
6
Apr 6, 2026
3.6LLM Judge Score
6
Mar 26, 2026
3.7LLM Judge Score
6
Mar 26, 2026
4.2LLM Judge Score
6
Mar 26, 2026
70.1Win Rate vs Confidence
6
Feb 26, 2026
36.7Lexical Coverage
4
Jun 2, 2026
1.67Vendi Score
4
Apr 29, 2026
93.81Length-Controlled Win Rate
4
Mar 26, 2026
100Pass Rate
4
Mar 20, 2026