List your company

Common-sense Reasoning Average

90.4Average AccuracyFRAME
53.209662.864872.5282.1752Mar 2, 2026Mar 22, 2026Apr 11, 2026May 1, 2026May 21, 2026Jun 10, 2026Jun 30, 2026

Evaluation results

MethodLinks
FRAMEBackbone=Qwen2.5, Model Scale=14B2026.0690.4
LoRABackbone=Qwen2.5, Model Scale=14B2026.0688.1
FRAMEBackbone=LLaMA-3.1, Model Scale=8B2026.0687.1
LoRABackbone=LLaMA-3.1, Model Scale=8B2026.0684.5
FRAMEBackbone=Qwen2.5, Model Scale=3B2026.0681.7
LoRABackbone=Qwen2.5, Model Scale=3B2026.0678.6
FRAMEBackbone=Qwen2.5, Model Scale=1.5B2026.0673.6
FP16Bit=16, Backbone=Qwen2.5-14B-Ins2026.0573.44
ProjQBit=3, Backbone=Qwen2.5-14B-Ins2026.0571.61
FP16Bit=16, Backbone=LLaMA2-13B2026.0570.61
LoRABackbone=Qwen2.5, Model Scale=1.5B2026.0670.2
ProjQBit=3, Backbone=LLaMA2-13B2026.0569.91
LoftQBit=3, Backbone=Qwen2.5-14B-Ins2026.0569.89
FP16Bit=16, Backbone=Qwen2.5-7B-Ins2026.0569.84
LoftQBit=3, Backbone=LLaMA2-13B2026.0569.1
ProjQBit=3, Backbone=Qwen2.5-7B-Ins2026.0568.33
FP16Bit=16, Backbone=LLaMA2-7B2026.0568.07
ProjQBit=3, Backbone=LLaMA2-7B2026.0567.13
QLoRA(RTN)Bit=3, Backbone=LLaMA2-7B2026.0566.56
LoftQBit=3, Backbone=LLaMA2-7B2026.0566.38
QLoRA(GPTQ)Bit=3, Backbone=LLaMA2-7B2026.0566.15
QLoRA(GPTQ)Bit=3, Backbone=Qwen2.5-7B-Ins2026.0565.08
QLoRA(RTN)Bit=3, Backbone=Qwen2.5-7B-Ins2026.0564.91
LoftQBit=3, Backbone=Qwen2.5-7B-Ins2026.0564.67
ProjQBit=2, Backbone=LLaMA2-13B2026.0561.84
LoftQBit=2, Backbone=LLaMA2-13B2026.0560.25
ProjQBit=2, Backbone=Qwen2.5-7B-Ins2026.0559.88
ProjQBit=2, Backbone=Qwen2.5-14B-Ins2026.0559.85
QLoRA(GPTQ)Bit=2, Backbone=Qwen2.5-7B-Ins2026.0559.66
LoftQBit=2, Backbone=Qwen2.5-14B-Ins2026.0559.08
LoftQBit=2, Backbone=Qwen2.5-7B-Ins2026.0558.86
MLRA-4Evaluation Mode=Zero-shot2026.0358.84
MLAEvaluation Mode=Zero-shot2026.0358.75
MLRA-2Evaluation Mode=Zero-shot2026.0358.68
MFAEvaluation Mode=Zero-shot2026.0358.53
GLA-2Evaluation Mode=Zero-shot2026.0358.3
GLA-4Evaluation Mode=Zero-shot2026.0358.19
GQAEvaluation Mode=Zero-shot2026.0357.89
TPAEvaluation Mode=Zero-shot2026.0357.87
MHAEvaluation Mode=Zero-shot2026.0357.81
GTAEvaluation Mode=Zero-shot2026.0357.8
ProjQBit=2, Backbone=LLaMA2-7B2026.0557.59
MQAEvaluation Mode=Zero-shot2026.0357.55
QLoRA(GPTQ)Bit=2, Backbone=LLaMA2-7B2026.0557.06
QLoRA(RTN)Bit=2, Backbone=Qwen2.5-7B-Ins2026.0556.64
LoftQBit=2, Backbone=LLaMA2-7B2026.0554.78
QLoRA(RTN)Bit=2, Backbone=LLaMA2-7B2026.0554.64