LIMI：简约之道，赋能代理

摘要

我们将“智能代理能力”定义为AI系统作为自主代理涌现出的能力，这种能力使其能够主动发现问题、提出假设，并通过与环境及工具的自发互动执行解决方案。这一核心能力标志着“AI代理时代”的开启，其背后是行业的一个关键转变：迫切需要AI系统不仅能思考，更能实际工作。尽管当前AI在推理和生成响应方面表现出色，但各行业更期待能执行任务、操作工具并推动现实成果的自主代理。随着代理智能成为区分认知系统与生产性工作者的关键特征，高效培养机器自主性变得至关重要。现有方法沿袭语言模型的传统扩展定律，认为更多数据能带来更好的代理能力。我们从根本上挑战这一范式。LIMI（少即是多，智能代理之道）证明，代理能力遵循截然不同的发展原则。通过战略性地聚焦于协作软件开发与科研工作流，我们展示了复杂的代理智能可以从少量但精心策划的自主行为示范中涌现。仅使用78个精心设计的训练样本，LIMI在综合代理能力基准测试中取得了73.5%的成绩，显著超越了当前最先进的模型：Kimi-K2-Instruct（24.1%）、DeepSeek-V3.1（11.9%）、Qwen3-235B-A22B-Instruct（27.5%）和GLM-4.5（45.1%）。尤为引人注目的是，LIMI相比使用10,000个样本训练的模型提升了53.7%的性能，以128倍少的样本实现了更优的代理智能。我们的研究确立了“代理效率原则”：机器自主性并非源于数据量的堆积，而是源于对高质量代理示范的战略性筛选。

English

We define Agency as the emergent capacity of AI systems to function as autonomous agents actively discovering problems, formulating hypotheses, and executing solutions through self-directed engagement with environments and tools. This fundamental capability marks the dawn of the Age of AI Agency, driven by a critical industry shift: the urgent need for AI systems that don't just think, but work. While current AI excels at reasoning and generating responses, industries demand autonomous agents that can execute tasks, operate tools, and drive real-world outcomes. As agentic intelligence becomes the defining characteristic separating cognitive systems from productive workers, efficiently cultivating machine autonomy becomes paramount. Current approaches assume that more data yields better agency, following traditional scaling laws from language modeling. We fundamentally challenge this paradigm. LIMI (Less Is More for Intelligent Agency) demonstrates that agency follows radically different development principles. Through strategic focus on collaborative software development and scientific research workflows, we show that sophisticated agentic intelligence can emerge from minimal but strategically curated demonstrations of autonomous behavior. Using only 78 carefully designed training samples, LIMI achieves 73.5% on comprehensive agency benchmarks, dramatically outperforming state-of-the-art models: Kimi-K2-Instruct (24.1%), DeepSeek-V3.1 (11.9%), Qwen3-235B-A22B-Instruct (27.5%), and GLM-4.5 (45.1%). Most strikingly, LIMI demonstrates 53.7% improvement over models trained on 10,000 samples-achieving superior agentic intelligence with 128 times fewer samples. Our findings establish the Agency Efficiency Principle: machine autonomy emerges not from data abundance but from strategic curation of high-quality agentic demonstrations.

LIMI：简约之道，赋能代理

LIMI: Less is More for Agency

摘要

Support