Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Open-ended Generation on WildBench (test)
Loading...
64.4
WildBench Score
Qwen3
8.864
23.282
37.7
52.118
Mar 6, 2026
WildBench Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
WildBench Score
Qwen3
Sparsity=Full, Model=Q...
2026.03
64.4
REAP
Sparsity=25%, Allocati...
2026.03
56.5
REAP
Sparsity=25%, Allocati...
2026.03
53.3
EAN
Sparsity=25%, Allocati...
2026.03
51.7
SEER
Sparsity=25%, Allocati...
2026.03
48.2
Freq
Sparsity=25%, Allocati...
2026.03
46.3
SEER
Sparsity=25%, Allocati...
2026.03
44.9
Freq
Sparsity=25%, Allocati...
2026.03
44.6
EAN
Sparsity=25%, Allocati...
2026.03
43.9
REAP
Sparsity=50%, Allocati...
2026.03
29.9
REAP
Sparsity=50%, Allocati...
2026.03
26.7
EAN
Sparsity=50%, Allocati...
2026.03
24.3
EAN
Sparsity=50%, Allocati...
2026.03
23.1
Freq
Sparsity=50%, Allocati...
2026.03
18.2
SEER
Sparsity=50%, Allocati...
2026.03
14.2
SEER
Sparsity=50%, Allocati...
2026.03
11.2
Freq
Sparsity=50%, Allocati...
2026.03
11
Feedback
Search any
task
Search any
task