Our new X account is live! Follow @wizwand_team for updates

Mathematical Reasoning on DeepMind-Mathematics (test)

64.1Accuracy

DART-Math-Llama3-70B (Uniform)

Updated 4d ago

Evaluation Results

Method	Links
DART-Math-Llama3-70B (Uniform) 2024.06		64.1
DART-Math-Llama3-70B (Prop2Diff) 2024.06		64.1
Llama3-70B-VRT 2024.06		62.8
DART-Math-DSMath-7B (Prop2Diff) 2024.06		61.6
Llama3-70B-MMIQC 2024.06		60.4
DeepSeekMath-7B-VRT 2024.06		60.2
DART-Math-DSMath-7B (Uniform) 2024.06		60.2
Llama2-70B-Xwin-Math-V1.1+ 2024.06		58
Llama3-70B-MetaMath 2024.06		53.2
DeepSeekMath-7B-MMIQC 2024.06		52.9
DeepSeekMath-7B-Instruct 2024.06		52.2
Llama3-70B-ICL 2024.06		51.7
DART-Math-Llama3-8B (Uniform) 2024.06		48.2
DART-Math-Llama3-8B (Prop2Diff) 2024.06		48
DeepSeekMath-7B-ICL 2024.06		45.2
DART-Math-Mistral-7B (Prop2Diff) 2024.06		45.1
Llama2-7B-Xwin-Math-V1.1+ 2024.06		43
DART-Math-Mistral-7B (Uniform) 2024.06		42
Llama3-8B-VRT 2024.06		41.7
Llama3-8B-MMIQC 2024.06		41
Mistral-7B-WizardMath-V1.1 (RL) 2024.06		38.4
Mistral-7B-MMIQC 2024.06		38
Mistral-7B-VRT 2024.06		35.6
Llama3-8B-MetaMath 2024.06		35
Mistral-7B-MetaMath 2024.06		28
Llama3-8B-ICL 2024.06		27.4
Mistral-7B-ICL 2024.06		23.5