Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Language Modeling on Perplexity Evaluation (zero-shot)
Loading...
10.98
PPL (zero-shot)
Dense
10.1124
15.9687
21.825
27.6813
Dec 8, 2025
PPL (zero-shot)
Updated 4d ago
Evaluation Results
Method
Method
Links
PPL (zero-shot)
Dense
Pruning Ratio=0%, Base...
2025.12
10.98
PP
Pruning Ratio=20%, Bas...
2025.12
12.52
PP
Pruning Ratio=25%, Bas...
2025.12
13.32
Token Filtering
Pruning Ratio=20%, Bas...
2025.12
13.37
SlimGPT w/o
Pruning Ratio=20%, Bas...
2025.12
13.8
FLAP
Pruning Ratio=20%, Bas...
2025.12
14.13
Token Filtering
Pruning Ratio=25%, Bas...
2025.12
14.69
SlimGPT w/o
Pruning Ratio=25%, Bas...
2025.12
15.1
FLAP
Pruning Ratio=25%, Bas...
2025.12
15.49
PP
Pruning Ratio=33%, Bas...
2025.12
15.83
Token Filtering
Pruning Ratio=33%, Bas...
2025.12
16.39
FLAP
Pruning Ratio=33%, Bas...
2025.12
17.79
SlimGPT w/o
Pruning Ratio=33%, Bas...
2025.12
18.11
PP
Pruning Ratio=50%, Bas...
2025.12
28.86
Token Filtering
Pruning Ratio=50%, Bas...
2025.12
29.22
FLAP
Pruning Ratio=50%, Bas...
2025.12
29.45
SlimGPT w/o
Pruning Ratio=50%, Bas...
2025.12
32.67
Feedback
Search any
task
Search any
task