MathΣtral | Mistral AI
Jul 16, 2024
Summary
Introduces MathΣtral, a specialized 7B parameter model derived from Mistral 7B, focused on advanced mathematical and STEM reasoning. Achieves state-of-the-art performance in its size category on benchmarks like MATH (56.6%) and MMLU (63.47%). Demonstrates improved reasoning with inference-time computation, scoring 68.37% on MATH with majority voting and 74.59% with a strong reward model. Highlights the performance/speed tradeoffs of purpose-built models and provides weights for use and fine-tuning.