List your company
ResearchFormerly WizWand

Mathematical Reasoning on CARP

53.6AccuracyGPT-4
8.56820.25931.9543.641May 23, 2024
Updated 8mo ago

Evaluation results

MethodLinks
GPT-42024.0553.6
Qwen-1.5-110BModel Scale=110B2024.0553.6
Qwen-1.5-72BModel Scale=72B2024.0553
JiuZhang3.0-8x7BModel Scale=8x7B2024.0552.3
DeepSeekMath-7B-RLModel Scale=7B2024.0551.6
JiuZhang3.0-7BModel Scale=7B, Backbone=Mistral-7B2024.0551
JiuZhang3.0-8BModel Scale=8B, Backbone=LLaMA-3-8B2024.0550.9
MAmmoTH2-8x7B-PlusModel Scale=8x7B2024.0545.8
DeepSeekMath-7B-InstructModel Scale=7B2024.0545.8
MAmmoTH2-8B-PlusModel Scale=8B2024.0544.8
DeepSeekMath-7BModel Scale=7B2024.0544.4
MAmmoTH2-7B-PlusModel Scale=7B2024.0544.3
Intern-Math-20BModel Scale=20B2024.0542.3
ChatGPT2024.0541.3
Mixtral-8x7BModel Scale=8x7B2024.0538.8
Qwen-1.5-7BModel Scale=7B2024.0538.6
Rho-1-Math-7BModel Scale=7B2024.0536.7
Llemma-34BModel Scale=34B2024.0536.5
Mistral-7B-MMIQCModel Scale=7B2024.0536.5
Abel-7B-002Model Scale=7B2024.0533.2
Math-Shepherd-Mistral-7BModel Scale=7B2024.0532.9
WizardMath-7B-1.1Model Scale=7B2024.0531.9
Llemma-7BModel Scale=7B2024.0531.8
Gemma-7BModel Scale=7B2024.0530.5
MetaMath-Mistral-7BModel Scale=7B2024.0530.5
LLAMA-3-8BModel Scale=8B2024.0529.2
InternLM-Math-7BModel Scale=7B2024.0528
Mistral-7BModel Scale=7B2024.0514.9
DeepSeek-7BModel Scale=7B2024.0510.3