List your company
ResearchFormerly WizWand

Reasoning Accuracy on BBH

76.7Accuracy (BBH)Qurating
35.630446.292756.95567.6173May 18, 2026May 24, 2026May 30, 2026Jun 5, 2026Jun 11, 2026Jun 17, 2026Jun 24, 2026
Updated 4mo ago

Evaluation results

MethodLinks
QuratingBase Model=OpenR1-Distill-7B2026.0676.7
Random BaselineBase Model=OpenR1-Distill-7B2026.0676.52
LESSBase Model=OpenR1-Distill-7B2026.0676.15
DRIFTBase Model=OpenR1-Distill-7B2026.0676.11
OpenR1-Distill-7BBase Model=OpenR1-Distill-7B2026.0676.04
RDSBase Model=OpenR1-Distill-7B2026.0675.88
BM25Base Model=OpenR1-Distill-7B2026.0675.76
IF (Impl. w. GraSS)Base Model=OpenR1-Distill-7B2026.0675.46
DSIRBase Model=OpenR1-Distill-7B2026.0673.86
iLLaDA 8BModel Architecture=Diffusion, Training tokens=12T2026.0671.3
Qwen2.5 7BModel Architecture=AR, Training tokens=18T2026.0663.9
Self-DistillationBase Model=OpenR1-Distill-7B2026.0660.82
Dream 7BModel Architecture=Diffusion, Training tokens=18T + 0.6T2026.0657.9
K2V-7B-QwenSetting=zero-shot, Temperature=0.6, Evaluations=Average of 4, Algorithm=DAPO2026.0555.36
LLaDA 8BModel Architecture=Diffusion, Training tokens=2.3T2026.0649.7
Qwen2.5-7B-InstructSetting=zero-shot, Temperature=0.6, Evaluations=Average of 42026.0549.63
DRIFTBase Model=Olmo3-7B-Instruct-SFT2026.0645.61
BM25Base Model=Olmo3-7B-Instruct-SFT2026.0644.88
Olmo3-7B-Instruct-SFTBase Model=Olmo3-7B-Instruct-SFT2026.0644.63
QuratingBase Model=Olmo3-7B-Instruct-SFT2026.0644.49
IF (Impl. w. GraSS)Base Model=Olmo3-7B-Instruct-SFT2026.0644.49
Random BaselineBase Model=Olmo3-7B-Instruct-SFT2026.0644.15
Self-DistillationBase Model=Olmo3-7B-Instruct-SFT2026.0643.89
RDSBase Model=Olmo3-7B-Instruct-SFT2026.0643.62
LESSBase Model=Olmo3-7B-Instruct-SFT2026.0643.48
K2V-3B-QwenSetting=zero-shot, Temperature=0.6, Evaluations=Average of 4, Algorithm=DAPO2026.0542.64
DSIRBase Model=Olmo3-7B-Instruct-SFT2026.0640.18
Qwen2.5-3B-InstructSetting=zero-shot, Temperature=0.6, Evaluations=Average of 42026.0537.21