List your company
ResearchFormerly WizWand

General Language Understanding on MMLU (Score)

73.59MMLU ScoreOrigin
21.922835.336448.7562.1636Feb 25, 2026Mar 5, 2026Mar 13, 2026Mar 22, 2026Mar 30, 2026Apr 7, 2026Apr 16, 2026
Updated 6mo ago

Evaluation results

MethodLinks
OriginBase Model=Qwen2.5-7B-Instruct2026.0273.59
SFTBase Model=Qwen2.5-7B-Instruct2026.0273.48
RLBase Model=Qwen2.5-7B-Instruct2026.0273.13
LARFTBase Model=Qwen2.5-7B-Instruct2026.0272.94
SFT+RLBase Model=Qwen2.5-7B-Instruct2026.0272.67
PositionIDBase Model=Qwen2.5-7B-Instruct2026.0272.04
RulerBase Model=Qwen2.5-7B-Instruct2026.0271.02
OriginBase Model=Llama-3.1-8B-Instruct2026.0268.45
SFTBase Model=Llama-3.1-8B-Instruct2026.0267.73
LARFTBase Model=Llama-3.1-8B-Instruct2026.0267.3
RLBase Model=Llama-3.1-8B-Instruct2026.0266.87
LARFTBase Model=Qwen2.5-3B-Instruct2026.0266.46
OriginBase Model=Qwen2.5-3B-Instruct2026.0265.07
SFT+RLBase Model=Llama-3.1-8B-Instruct2026.0265.05
SFTBase Model=Qwen2.5-3B-Instruct2026.0265
RLBase Model=Qwen2.5-3B-Instruct2026.0265
SFTBase Model=Llama-3.2-3B-Instruct2026.0261.74
OriginBase Model=Llama-3.2-3B-Instruct2026.0261.5
RLBase Model=Llama-3.2-3B-Instruct2026.0261.35
LARFTBase Model=Llama-3.2-3B-Instruct2026.0260.87
Base Model2026.0460.8
NGDiff2026.0460.8
MOLLM2026.0460.8
Upper Bound2026.0460.7
SFT+RLBase Model=Llama-3.2-3B-Instruct2026.0260.68
SimNPOTraining objective=Obj1 only2026.0460.4
Ensemble Teacher (ET)2026.0460.3
PositionIDBase Model=Qwen2.5-3B-Instruct2026.0259.97
MULE2026.0459.2
FLATTraining objective=Obj1 only2026.0458.92
SFT+RLBase Model=Qwen2.5-3B-Instruct2026.0254.75
SFT2026.0450.2
DUETTeacher model integration=Ensemble Teacher2026.0449.9
RulerBase Model=Qwen2.5-3B-Instruct2026.0241.94
PositionIDBase Model=Llama-3.1-8B-Instruct2026.0240.35
RulerBase Model=Llama-3.1-8B-Instruct2026.0237.15
GATraining objective=Obj1 only2026.0436.45
RulerBase Model=Llama-3.2-3B-Instruct2026.0224.05
PositionIDBase Model=Llama-3.2-3B-Instruct2026.0223.91