Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
INT2 Quantization on OPT-6.7B
Loading...
11.68
Memory Footprint (GB)
OmniQuant
11.344
13.612
15.88
18.148
Feb 14, 2024
Memory Footprint (GB)
Updated 4d ago
Evaluation Results
Method
Method
Links
Memory Footprint (GB)
OmniQuant
Target=attention-wise...
2024.02
11.68
OmniQuant
Target=attention-wise...
2024.02
11.68
aespa
Target=attention-wise...
2024.02
12.26
aespa
Target=attention-wise...
2024.02
12.26
OPTQ
Target=layer-wise reco...
2024.02
13.07
Z-FOLD
Target=layer-wise reco...
2024.02
13.07
OPTQ
Target=layer-wise reco...
2024.02
13.07
Z-FOLD
Target=layer-wise reco...
2024.02
13.07
AffineQuant
Target=attention-wise...
2024.02
20.08
AffineQuant
Target=attention-wise...
2024.02
20.08
Feedback
Search any
task
Search any
task