Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Long-context understanding on LongBench (Difficulty and Length Aggregation)

31.8Overall Average Score

BLASST

-1.12647.421815.9724.5182Dec 12, 2025Dec 21, 2025Dec 31, 2025Jan 9, 2026Jan 19, 2026Jan 28, 2026Feb 7, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2025.12
31.830.732.538.329.825
2025.12
31.429.732.538.328.825
2025.12
31.228.632.836.730.224.1
2025.12
30.629.231.538.32626.9
2025.12
25.728.823.824.426.526.2
2026.02
11.12-----
2026.02
10.81-----
2026.02
9.72-----
2026.02
9.71-----
2026.02
9.49-----
2026.02
8.31-----
2026.02
8.29-----
2026.02
6.07-----
2026.02
3.84-----
2026.02
3.64-----
2026.02
0.14-----
2026.02
0.14-----