Large Language Model Evaluation on LLaMA 1B 3.2 benchmark leaderboard · SOTA2 Research

Large Language Model Evaluation on LLaMA 1B 3.2

9.75Perplexity (PPL)baseline

Evaluation Results

Method	Links
baseline2026.03		9.75	54.82	36.76
SERQ2026.03		12.52	50.44	26.34
QuaRot2026.03		13.17	50.03	26.64
SpinQuant2026.03		13.47	48.95	26.38
SmoothQ(g128)2026.03		69.22	40.04	24.43
SmoothQ2026.03		153,000	35.59	24.37