Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Multi-path Speculative Decoding on Average across models (Qwen, Gemma, Llama) (test)
Loading...
16.69
Throughput (tokens/s)
SpecInfer NDE
14.7556
15.2578
15.76
16.2622
Feb 19, 2026
Throughput (tokens/s)
Updated 4d ago
Evaluation Results
Method
Method
Links
Throughput (tokens/s)
SpecInfer NDE
Expansion Mode=Neural...
2026.02
16.69
SpecTr NDE
Expansion Mode=Neural...
2026.02
16.4
NaiveTree NDE
Expansion Mode=Neural...
2026.02
16.18
Traversal
Expansion Mode=Static
2026.02
15.89
NSS NDE
Expansion Mode=Neural...
2026.02
15.05
Khisti NDE
Expansion Mode=Neural...
2026.02
14.83
Feedback
Search any
task
Search any
task