Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Language Modeling on Language Modeling (test)
Loading...
6.2
PPL
Qwen3-4B + FBS-Full (ours)
6.188
6.269
6.35
6.431
Jan 29, 2026
PPL
Latency (ms)
TFLOPs (rel)
Bypass Ratio (%)
Updated 4d ago
Evaluation Results
Method
Method
Links
PPL
Latency (ms)
TFLOPs (rel)
Bypass Ratio (%)
Qwen3-4B + FBS-Full (ours)
Backbone=Qwen3-4B, Mod...
2026.01
6.2
532
0.7
36
Qwen3-4B + FBS-S1
Backbone=Qwen3-4B, Mod...
2026.01
6.3
755
1.03
0
Qwen3-4B + EAGLE-2 (Group A)
Backbone=Qwen3-4B, Met...
2026.01
6.3
555
0.74
30
Qwen3-4B-Instruct (Baseline)
Backbone=Qwen3-4B, Var...
2026.01
6.4
760
1
0
Qwen3-4B + SpecDec (Group A)
Backbone=Qwen3-4B, Met...
2026.01
6.4
646
0.9
22
Qwen3-4B + Lookahead (Group A)
Backbone=Qwen3-4B, Met...
2026.01
6.4
595
0.82
15
Qwen3-4B + Medusa (Group A)
Backbone=Qwen3-4B, Met...
2026.01
6.5
570
0.8
18
Feedback
Search any
task
Search any
task