Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Inference Throughput on Qwen3-8B
Loading...
26.74
Throughput (tokens/s)
ROCKET
24.2128
24.8689
25.525
26.1811
Feb 11, 2026
Throughput (tokens/s)
Updated 4d ago
Evaluation Results
Method
Method
Links
Throughput (tokens/s)
ROCKET
Compression Ratio=20%,...
2026.02
26.74
ROCKET
Compression Ratio=30%,...
2026.02
26.6
ROCKET
Compression Ratio=40%,...
2026.02
26.36
CoSpaDi
Compression Ratio=30%,...
2026.02
25.76
CoSpaDi
Compression Ratio=40%,...
2026.02
25.62
CoSpaDi
Compression Ratio=20%,...
2026.02
25.45
SVD-LLM
Compression Ratio=40%,...
2026.02
24.72
SVD-LLM
Compression Ratio=20%,...
2026.02
24.36
SVD-LLM
Compression Ratio=30%,...
2026.02
24.31
Feedback
Search any
task
Search any
task