Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
INT2 Quantization on LLaMA 7B
Loading...
8.76
Memory Cost (GB)
OPTQ
8.1392
12.3296
16.52
20.7104
Feb 14, 2024
Memory Cost (GB)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Memory Cost (GB)
OPTQ
Target=layer-wise reco...
2024.02
8.76
Z-FOLD
Target=layer-wise reco...
2024.02
8.76
OPTQ
Target=layer-wise reco...
2024.02
8.76
Z-FOLD
Target=layer-wise reco...
2024.02
8.76
OmniQuant
Target=attention-wise...
2024.02
12.61
OmniQuant
Target=attention-wise...
2024.02
12.61
aespa
Target=attention-wise...
2024.02
21.69
aespa
Target=attention-wise...
2024.02
21.69
AffineQuant
Target=attention-wise...
2024.02
24.28
AffineQuant
Target=attention-wise...
2024.02
24.28
Feedback
Search any
task
Search any
task