Astral-Math-v1: Multi-Model TIR Mathematical Reasoning Dataset

A large-scale collection of mathematical problems categorized for Multi-Model TIR tasks. It provides structured data for training and evaluating reasoning-based mathematical solvers through multi-step logic.

Use Cases

• Train models to solve mathematical problems using TIR sequences
• Benchmark multi-model reasoning performance
• Improve intermediate reasoning steps in math solvers

Strengths

• Features Multi-Model TIR data structures
• Focuses on mathematical problem solving
• Large-scale dataset for multi-model training