Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Language Modeling on Qwen3-0.6B (val)
Loading...
31.45
Validation Perplexity
Delta Block
31.4128
31.6639
31.915
32.1661
May 13, 2026
Validation Perplexity
Throughput (Tok/s)
Memory Usage (GB)
Updated 14d ago
Evaluation Results
Method
Method
Links
Validation Perplexity
Throughput (Tok/s)
Memory Usage (GB)
Delta Block
L (Layers)=28, N (Rout...
2026.05
31.45
117
25.2
Baseline
L (Layers)=28, Steps=1...
2026.05
32.22
173
13.8
AttnRes
L (Layers)=28, N (Rout...
2026.05
32.38
111
28.4
Feedback
Search any
task
Search any
task