List your company
ResearchFormerly WizWand

Mathematical Reasoning on TheoremQA

51.2AccuracyHIPPO
3.536815.910928.28540.6591Jun 18, 2024Oct 19, 2024Feb 19, 2025Jun 23, 2025Oct 24, 2025Feb 24, 2026Jun 28, 2026
Updated 4mo ago

Evaluation results

MethodLinks
HIPPOTraining Data=DeepScaleR2026.0651.2
Qwen3.5-9BShots=52026.0449.25
Qwen2-72B2024.0743.1
Qwen3-14BShots=52026.0442.88
XekRung-8BShots=52026.0441.63
Qwen3-8BShots=52026.0440.12
Llama-3.3-70B-InstructShots=52026.0438.88
Mixtral-8x22B2024.0735.9
DeepSeekMath-7B-RLTraining Method=RL, Backbone=DeepSeekMath-7B2024.0635.9
Qwen1.5-110B2024.0734.9
Qwen2-57B-A14BArchitecture=MoE, # Act Params=14B, # Params=57B2024.0733.5
DART-Math-DSMath-7BTraining Method=SFT, Data Selection Strategy=Uniform, Backbone=DeepSeekMath-7B2024.0632.5
Llama-3-70B2024.0732.3
DeepSeekMath-7B-DART-MathBase Model=DeepSeekMath-7B, # Samples=590K2025.0332.2
DART-Math-DSMath-7BTraining Method=SFT, Data Selection Strategy=Prop2Diff, Backbone=DeepSeekMath-7B2024.0632.2
Qwen3-4B + SFTTraining Data=DeepScaleR2026.0631.8
Qwen1.5-72B2024.0729.3
Qwen3-4BTraining Data=DeepScaleR2026.0629.1
Qwen1.5-32BArchitecture=Dense, # Act Params=34B, # Params=34B2024.0728.8
DeepSeekMath-7B-InstructBase Model=DeepSeekMath-7B, # Samples=780K2025.0328.1
DeepSeekMath-7B-DART-Math†Base Model=DeepSeekMath-7B, # Samples=60K2025.0327.4
DeepSeekMath-7B-RFTBase Model=DeepSeekMath-7B, # Samples=590K2025.0327.2
MathFusion-DSMath-7BBase Model=DeepSeekMath-7B, # Samples=195K2025.0327
Llama-3.1-8B-InstructShots=52026.0426
MathFusion-DSMath-7BBase Model=DeepSeekMath-7B, # Samples=60K2025.0324.6
MathFusion-DSMath-7BBase Model=DeepSeekMath-7B, # Samples=30K, Training Strategy=Parallel2025.0323.8
SecGPT-14BShots=52026.0423.5
DeepSeekMath-7B-MMIQCBase Model=DeepSeekMath-7B, # Samples=2.3M2025.0323.4
Mixtral-8x7BArchitecture=MoE, # Act Params=12B, # Params=47B2024.0723.2
Llama3-8B-DART-Math†Base Model=Llama3-8B, # Samples=60K2025.0322.9
MathFusion-DSMath-7BBase Model=DeepSeekMath-7B, # Samples=30K, Training Strategy=Sequential2025.0322.8
Foundation-Sec-8B-ReasoningShots=52026.0422.25
MathFusion-Llama3-8BBase Model=Llama3-8B, # Samples=60K2025.0320
Llama3-8B-DART-MathBase Model=Llama3-8B, # Samples=590K2025.0319.4
MathFusion-DSMath-7BBase Model=DeepSeekMath-7B, # Samples=30K, Training Strategy=Conditional2025.0319
DeepSeekMath-7B-MetaMathBase Model=DeepSeekMath-7B, # Samples=60K2025.0318.9
MathFusion-Llama3-8BBase Model=Llama3-8B, # Samples=30K, Training Strategy=Parallel2025.0318.9
Mistral-7B-DART-Math†Base Model=Mistral-7B, # Samples=60K2025.0318.2
MathFusion-Mistral-7BBase Model=Mistral-7B, # Samples=60K2025.0318.1
Mistral-7B-DART-MathBase Model=Mistral-7B, # Samples=590K2025.0317
MathFusion-Llama3-8BBase Model=Llama3-8B, # Samples=30K, Training Strategy=Sequential2025.0317
Mistral-7B-WizardMath-V1.1Base Model=Mistral-7B, # Samples=418K2025.0316.6
Mistral-7B-RFTBase Model=Mistral-7B, # Samples=590K2025.0316.2
Mistral-7B-MMIQCBase Model=Mistral-7B, # Samples=2.3M2025.0316.2
Llama3-8B-MMIQCBase Model=Llama3-8B, # Samples=2.3M2025.0316.2
Llama3-8B-MetaMath†Base Model=Llama3-8B, # Samples=60K2025.0316.1
MathFusion-Mistral-7BBase Model=Mistral-7B, # Samples=30K, Training Strategy=Sequential2025.0315.5
MathFusion-Llama3-8BBase Model=Llama3-8B, # Samples=30K, Training Strategy=Conditional2025.0315.5
MathFusion-Mistral-7BBase Model=Mistral-7B, # Samples=30K, Training Strategy=Parallel2025.0315.2
Llama3-8B-RFTBase Model=Llama3-8B, # Samples=590K2025.0314.9
DeepSeekMath-7B-RefAugBase Model=DeepSeekMath-7B, # Samples=30K2025.0314.4
DeepSeekMath-7B-RefAugBase Model=DeepSeekMath-7B, # Samples=60K2025.0314
Mistral-7B-MetaMathBase Model=Mistral-7B, # Samples=400K2025.0314
Llama3-8B-MetaMathBase Model=Llama3-8B, # Samples=400K2025.0313.8
Llama3-8B-RefAugBase Model=Llama3-8B, # Samples=30K2025.0313.6
Llama3-8B-RefAugBase Model=Llama3-8B, # Samples=60K2025.0313
MathFusion-Mistral-7BBase Model=Mistral-7B, # Samples=30K, Training Strategy=Conditional2025.0312.8
Mistral-7B-MetaMath†Base Model=Mistral-7B, # Samples=60K2025.0312.2
Mistral-7B-RefAugBase Model=Mistral-7B, # Samples=60K2025.0311.1
DeepSeekMath-7B-StandardBase Model=DeepSeekMath-7B, # Samples=15K2025.0311
Mistral-7B-RefAugBase Model=Mistral-7B, # Samples=30K2025.0311
Llama3-8B-StandardBase Model=Llama3-8B, # Samples=15K2025.0310.9
Llama3-8B-MMIQC†Base Model=Llama3-8B, # Samples=60K2025.0310.6
Mistral-7B-StandardBase Model=Mistral-7B, # Samples=15K2025.037.6
DeepSeekMath-7B-MMIQC†Base Model=DeepSeekMath-7B, # Samples=60K2025.036.8
Mistral-7B-MMIQC†Base Model=Mistral-7B, # Samples=60K2025.035.9
Llama-Primus-Reasoning-8BShots=52026.045.37