OpenMath2-Llama3.1-70B

Reported on 5 benchmarks across 5 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Reasoning2 results

Math Word Problem SolvingonMATH
Accuracy· uses extra data· 2024-10-02
71.9
best: 89.7 (Gemini 2.0 Flash Experimental)
OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Data arXiv:2410.01560
Arithmetic ReasoningonGSM8K
Accuracy· uses extra data· 2024-10-02
94.9
best: 97.72 (Claude 3.5 Sonnet (HPT))
OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Data arXiv:2410.01560

Knowledge Base2 results

Mathematical Question AnsweringonMATH
Accuracy· uses extra data· 2024-10-02
71.9
best: 89.7 (Gemini 2.0 Flash Experimental)
OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Data arXiv:2410.01560
Mathematical ReasoningonMATH
Accuracy· uses extra data· 2024-10-02
71.9
best: 89.7 (Gemini 2.0 Flash Experimental)
OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Data arXiv:2410.01560

Natural Language Processing1 result

Question AnsweringonMATH
Accuracy· uses extra data· 2024-10-02
71.9
best: 89.7 (Gemini 2.0 Flash Experimental)
OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Data arXiv:2410.01560