优化AI调度层可大幅降低企业智能体的耗能与成本，效果超越换模型。
The Harness Effect: How Orchestration Design Sets the Token Economics of Enterprise Agentic AI
- 用固定模型对比不同调度设计，发现调度层决定效率核心
- 同一任务下成本降41%，耗时减44%，令牌数少38%
- 调度层提升性价比，适合所有规模企业部署智能体系统
当前企业级智能体开发依赖于‘令牌堆叠’：通过增加推理长度、交互轮次、工具调用和上下文规模来换取能力，导致单任务令牌消耗增速远超任务价值。尽管单令牌价格下降，总支出仍持续上升。本文提出关键杠杆是‘调度层’（harness）——负责组装上下文、暴露工具、安排轮次、分配任务，并支持可观测性与治理。在22个固定评估任务上，使用六种基础模型（Claude Sonnet 4.6, Gemini 3.1, Gemini Flash 3.5, Qwen 3.6, GLM 5.1, Palmyra X6），仅替换调度层（传统生产流程 vs. Writer Agent Harness），结果显示：在模型不变的情况下，调度层使混合成本下降41%（$0.21→$0.12）、中位运行时间减少44%（48s→27s）、每任务令牌数降低38%（14.2k→8.8k），任务完成质量保持一致（0.78→0.81，样本量下方向性显著）。该效率提升对所有模型通用（每模型成本降33%-61%），而质量增益与模型基线能力高度相关（r=0.99, n=6），称为“调度杠杆”。单位美元产出质量提升82%，每百万令牌完成任务数从54.9增至92.0。在此工作负载下，调度层对成本的影响超过整个模型选型范围。本文还形式化了调度层的令牌经济模型（含提示缓存下的有效输入价格），总结六大机制家族（如缓存形状规范至失败支出治理），并横向比较六种主流智能体系统。结论是：调度层是唯一能在组织所用所有模型上实现效率倍增的组件，涵盖现有与未来模型。
原文摘要 · Abstract (English)
Agentic AI development today runs on token maxing: buying capability with tokens -- longer reasoning traces, more turns, wider tool payloads, bigger replayed contexts -- so tokens per task grow faster than task value. Falling per-token prices mask the pattern; total spend rises anyway. We argue the decisive lever against token maxing is the harness: the orchestration layer that assembles context, exposes tools, sequences turns, delegates work, and carries enterprise observability and governance. We isolate it with a controlled swap: 22 locked evaluation tasks, six foundation models (Claude Sonnet 4.6, Gemini 3.1, Gemini Flash 3.5, Qwen 3.6, GLM 5.1, Palmyra X6), changing only the orchestration layer -- a frozen conventional production loop versus the Writer Agent Harness. Holding models constant, the harness cuts blended cost per task 41% ($0.21->$0.12), median wall-clock 44% (48s->27s), and tokens per task 38% (14.2k->8.8k), with task-completion quality at parity (0.78->0.81, directional at this sample size). Efficiency is model-invariant -- every model gets cheaper (33-61%) -- while quality gains are capability-dependent: a model's gain correlates almost perfectly with its baseline strength (r=0.99, n=6), a phenomenon we term harness leverage. Quality per dollar rises 82%; task-completions per million tokens rise from 54.9 to 92.0. On this workload the orchestration layer moved cost per task more than the full spread of the model menu did. We formalize token economics at the orchestration layer (including effective input price under prompt caching), detail the six mechanism families behind the effect -- cache-shape discipline to failure-spend governance -- compare six widely used agent systems on the same axes, and argue the harness is the one component whose efficiency multiplies across every model an organization runs -- present and future.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。