Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Large Language Model Evaluation on LLaMA 3B 3.2
Loading...
7.81
PPL
baseline
-1,483.8776
8,585.0137
18,653.905
28,722.7963
Mar 9, 2026
PPL
Zero-Shot Performance
MMLU Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
PPL
Zero-Shot Performance
MMLU Accuracy
baseline
#Bits=FP16
2026.03
7.81
62.66
54.06
SERQ
#Bits=W4A4
2026.03
9.15
58.5
45.7
QuaRot
#Bits=W4A4
2026.03
9.73
55.76
44.75
SpinQuant
#Bits=W4A4
2026.03
10.15
56.88
42.42
SmoothQ(g128)
#Bits=W4A4
2026.03
53.33
44.17
27.31
SmoothQ
#Bits=W4A4
2026.03
37,300
35.72
23.41
Feedback
Search any
task
Search any
task
