# Synthetic data distillation

Phi-4-mini-reasoning is a lightweight open-source model focused on mathematical reasoning, supporting a context length of 128K tokens and suitable for environments with limited computing resources.

NeuralHermes is a large language model based on OpenHermes-2.5-Mistral-7B, further fine-tuned through Direct Preference Optimization (DPO), demonstrating excellent performance across multiple benchmarks.