AI code generation

This GitHub repository contains the training and inference code used to replicate our winning solution in the AI ​​Math Olympiad (AIMO) Progress Award 1. Our solution consists of four main parts: a recipe for fine-tuning DeepSeekMath-Base 7B to solve mathematical problems using Tool Integrated Reasoning (TIR); two high-quality training datasets of ~1 million mathematical problems and solutions; a self-consistent decoding algorithm for generating solution candidates with code execution feedback (SC-TIR); and four carefully selected validation sets from AMC, AIME, and MATH to guide model selection and avoid overfitting to public leaderboards.