List your company

Multitask Language Understanding on MMLU (Score)

87.21MMLU ScoreTrinity Large Preview
54.678863.124471.5780.0156Feb 19, 2026

Evaluation results

MethodLinks
Trinity Large Previewtuning=instruct-tuned2026.0287.21
IProXCandidate Training Data=CoT, rho=0.32026.0256.96
Llama3.2-3BCandidate Training Data=BioInstruct2026.0256.61
Llama3.2-3BCandidate Training Data=CoT2026.0256.53
IProXCandidate Training Data=CoT, rho=0.52026.0256.48
IProXCandidate Training Data=CoT, rho=0.72026.0256.26
IProXCandidate Training Data=BioInstruct, rho=0.32026.0256.25
IProXCandidate Training Data=BioInstruct, rho=0.52026.0256.21
Llama3.2-1BCandidate Training Data=CoT2026.0256.17
IProXCandidate Training Data=BioInstruct, rho=0.72026.0256.09
Llama3.2-1BCandidate Training Data=BioInstruct2026.0255.93