At 30K and 60K data scales, AgenticMath outperforms all compared baselines across four base models on six math benchmarks¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person · recheck by 2028: tool and model accuracy changes quickly

Subclaims¶

q2 i? With 30K samples, AgenticMath-Qwen2.5-3B reaches 53.7 average accuracy, surpassing MathFusion by over 15 points. → Xianyang Liu 2025

Evidence¶

Xianyang Liu 2025¶

Xianyang Liu, Yilin Liu, Shuai Wang, Hao Cheng, Andrew Estornell, Yuzhi Zhao, Jun Shu, Jiaheng Wei. (2025). AgenticMath: Enhancing LLM Reasoning via Agentic-based Math Data Generation. arXiv preprint. https://arxiv.org/abs/2510.19361 doi:10.48550/arxiv.2510.19361

q2 · i? · design · r2

Evaluation across in-domain (MATH, GSM8K) and out-of-domain benchmarks (Table 1) at 30K and 60K scales on Qwen2.5-3B, DeepSeekMath-7B, Mistral-7B, and Llama3-8B; the text reports "AgenticMath-Qwen2.5-3B reaches 53.7 average accuracy" at 30K.

"with 30K samples, AgenticMath-Qwen2.5-3B reaches 53.7 average accuracy, surpassing MathFusion by over 15 points."

Discussion¶

Related Claims¶

• Increasing AgenticMath-generated training data from 7.5K to 22.5K samples yields consistent accuracy gains for Llama3-8B across benchmarks — related
• Under the same teacher model, AgenticMath outperforms prior data-generation methods, indicating gains arise from framework design rather than teacher strength — related
• Fine-tuning LLMs on 30K-90K AgenticMath samples matches or surpasses baselines trained on 400K-2.3M samples on math reasoning benchmarks — related