List your company
ResearchFormerly WizWand

Mathematical Reasoning on All Average

60AccuracyBudget Prompt
38.1643.8349.555.17May 9, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Budget PromptModel Scale=Qwen3-4B-Thinking-2507, # Data=–2026.056018,448
OriginalModel Scale=Qwen3-4B-Thinking-2507, # Data=–2026.0559.319,699
SFT-S1.1Model Scale=DeepSeek-R1-Distill-Qwen-32B, # Data=1K2026.0557.212,674
HINT TUNINGModel Scale=Qwen3-4B-Thinking-2507, # Data=1K2026.0556.412,994
SFT-S1.1Model Scale=DeepSeek-R1-Distill-Qwen-14B, # Data=1K2026.0551.813,344
OriginalModel Scale=DeepSeek-R1-Distill-Qwen-32B, # Data=–2026.0551.612,737
SFT-S1.1Model Scale=Qwen3-4B-Thinking-2507, # Data=1K2026.0551.411,911
HINT TUNINGModel Scale=DeepSeek-R1-Distill-Qwen-32B, # Data=1K2026.0550.99,242
Budget PromptModel Scale=DeepSeek-R1-Distill-Qwen-14B, # Data=–2026.0548.511,556
OriginalModel Scale=DeepSeek-R1-Distill-Qwen-14B, # Data=–2026.0548.312,671
Budget PromptModel Scale=DeepSeek-R1-Distill-Qwen-32B, # Data=–2026.0548.310,671
HINT TUNINGModel Scale=DeepSeek-R1-Distill-Qwen-14B, # Data=1K2026.05488,850
SFT-S1.1Model Scale=DeepSeek-R1-Distill-Qwen-7B, # Data=1K2026.0543.413,772
Budget PromptModel Scale=DeepSeek-R1-Distill-Qwen-7B, # Data=–2026.0540.912,916
OriginalModel Scale=DeepSeek-R1-Distill-Qwen-7B, # Data=–2026.0540.713,786
AdaptThinkModel Scale=DeepSeek-R1-Distill-Qwen-7B, # Data=40K†2026.0540.711,256
HINT TUNINGModel Scale=DeepSeek-R1-Distill-Qwen-7B, # Data=1K2026.0540.79,027
VeriThinkerModel Scale=DeepSeek-R1-Distill-Qwen-7B, # Data=340K2026.0540.510,463
EfficientRModel Scale=DeepSeek-R1-Distill-Qwen-7B, # Data=3.2K†, Alpha=0.12026.0539.411,527
EfficientRModel Scale=DeepSeek-R1-Distill-Qwen-7B, # Data=3.2K†, Alpha=0.42026.053910,950