Instella：全开放语言模型，展现卓越性能

摘要

大型语言模型（LLMs）已在广泛任务中展现出卓越性能，但多数高性能模型仍保持闭源或部分开放，限制了研究的透明度和可复现性。本研究推出Instella系列——一个完全基于开放数据和代码库训练的三百亿参数全开源语言模型家族。依托AMD Instinct MI300X GPU的算力支持，Instella通过大规模预训练、通用指令微调以及与人类偏好的对齐训练得以开发。尽管预训练词元数量显著少于同期多数模型，Instella在全开源模型中取得了领先性能，并与同规模的主流开放权重模型相媲美。我们进一步发布两个专用变体：支持128K词元长上下文处理的Instella-Long，以及通过数学任务监督微调和强化学习增强的推理专用模型Instella-Math。这些成果共同确立了Instella作为透明、高效、多用途社区替代方案的定位，推动了开放可复现语言模型研究的发展。

English

Large language models (LLMs) have demonstrated remarkable performance across a wide range of tasks, yet the majority of high-performing models remain closed-source or partially open, limiting transparency and reproducibility. In this work, we introduce Instella, a family of fully open three billion parameter language models trained entirely on openly available data and codebase. Powered by AMD Instinct MI300X GPUs, Instella is developed through large-scale pre-training, general-purpose instruction tuning, and alignment with human preferences. Despite using substantially fewer pre-training tokens than many contemporaries, Instella achieves state-of-the-art results among fully open models and is competitive with leading open-weight models of comparable size. We further release two specialized variants: Instella-Long, capable of handling context lengths up to 128K tokens, and Instella-Math, a reasoning-focused model enhanced through supervised fine-tuning and reinforcement learning on mathematical tasks. Together, these contributions establish Instella as a transparent, performant, and versatile alternative for the community, advancing the goal of open and reproducible language modeling research.

Instella：全开放语言模型，展现卓越性能

Instella: Fully Open Language Models with Stellar Performance

摘要

Support