Hermes 4 技术报告

摘要

我们推出了Hermes 4系列模型，这是一组融合了结构化多轮推理与广泛指令跟随能力的混合推理模型。本文详细阐述了在数据整理、合成、训练及评估过程中遇到的挑战，并概述了为大规模应对这些挑战所采用的解决方案。我们全面评估了模型在数学推理、编程、知识理解、阅读理解及对齐基准测试中的表现，既报告了量化性能，也进行了定性行为分析。为支持开放研究，所有模型权重均已公开发布于https://huggingface.co/collections/NousResearch/hermes-4-collection-68a731bfd452e20816725728。

English

We present Hermes 4, a family of hybrid reasoning models that combine structured, multi-turn reasoning with broad instruction-following ability. We describe the challenges encountered during data curation, synthesis, training, and evaluation, and outline the solutions employed to address these challenges at scale. We comprehensively evaluate across mathematical reasoning, coding, knowledge, comprehension, and alignment benchmarks, and we report both quantitative performance and qualitative behavioral analysis. To support open research, all model weights are published publicly at https://huggingface.co/collections/NousResearch/hermes-4-collection-68a731bfd452e20816725728

Hermes 4 技术报告

Hermes 4 Technical Report

摘要

Support