每激活皆增效：将通用推理器扩展至万亿级开放语言基础

摘要

我们推出Ling 2.0系列——一个基于"每次激活皆提升推理能力"原则构建的面向推理的语言基础模型。该系列采用统一的混合专家（MoE）范式，设计参数规模从数百亿至一万亿，重点实现高稀疏性、跨尺度一致性及基于经验缩放定律的高效性。系列包含三款非思维（指令）模型：Ling-mini-2.0、Ling-flash-2.0和Ling-1T，总参数量从160亿到1万亿，相比稠密模型最高可实现7倍激活计算效率。Ling 2.0融合了模型架构、预训练、后训练与基础设施的协同创新：采用支持MTP的高稀疏MoE实现高效推理，配备面向推理的数据与训练中程思维链激活，基于强化学习的微调（DFT、Evo-CoT），以及全尺度FP8训练与细粒度异构流水线。在万亿规模上，Ling-1T确立了推理精度与计算效率的新帕累托前沿，证明稀疏激活与推理目标精准对齐时，可实现可扩展的高效智能。整体而言，Ling 2.0为推进未来推理与思维模型（包括基于同架构的Ring系列）提供了连贯、开放且高效的基础框架。

English

We introduce Ling 2.0, a series reasoning-oriented language foundation built upon the principle that every activation boosts reasoning capability. Designed to scale from tens of billions to one trillion parameters under a unified Mixture-of-Experts (MoE) paradigm, Ling 2.0 emphasizes high sparsity, cross-scale consistency, and efficiency guided by empirical scaling laws. The series includes three non-thinking (instruct) models - Ling-mini-2.0, Ling-flash-2.0, and Ling-1T - ranging from 16B to 1T total parameters and achieving up to 7-fold active-compute efficiency compared with dense counterparts. Ling 2.0 integrates coordinated innovations across model architecture, pre-training, post-training, and infrastructure: a high-sparsity MoE with MTP for efficient reasoning, reasoning-oriented data and mid-training CoT activation, reinforcement-based fine-tuning (DFT, Evo-CoT), and full-scale FP8 training with fine-grained heterogeneous pipelines. At the trillion scale, Ling-1T establishes a new Pareto frontier of reasoning accuracy versus computational efficiency, demonstrating that sparse activation, when properly aligned with reasoning objectives, enables scalable and efficient intelligence. Collectively, Ling 2.0 provides a coherent, open, and efficient foundation for advancing future reasoning and thinking models, including the Ring series built upon the same base.

每激活皆增效：将通用推理器扩展至万亿级开放语言基础

Every Activation Boosted: Scaling General Reasoner to 1 Trillion Open Language Foundation

摘要

Support