Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Model Compression on Qwen 7B 2.5
Loading...
2.7
Model Size (GB)
UniQL
2.2
5.575
8.95
12.325
Dec 3, 2025
Model Size (GB)
Updated 3mo ago
Evaluation Results
Method
Method
Links
Model Size (GB)
UniQL
W-bit=4, Prun. p%=35%
2025.12
2.7
UniQL
W-bit=4, Prun. p%=0%
2025.12
3.9
TRT-AWQ
W-bit=4*, Prun. p%=0%
2025.12
5.6
TAO-HQQ
W-bit=4*, Prun. p%=0%
2025.12
6
MoDeGPT
W-bit=16, Prun. p%=15%
2025.12
13.2
SVD-LLM
W-bit=16, Prun. p%=15%
2025.12
13.3
FP16
W-bit=16, Prun. p%=0%
2025.12
15.2
Feedback
Search any
task
Search any
task
