Introducing Cursor Router介绍 Cursor Router
Today we're launching Cursor Router, our intelligent model router for teams and enterprises.今天，我们正式推出 Cursor Router，这是专为团队和企业打造的智能模型路由工具。
Cursor Router lets teams automatically route every request to the most capable model for the task, delivering frontier intelligence at a lower cost.Cursor Router 能够让团队自动为每项任务分配最合适的模型，在降低成本的同时提供顶尖的智能表现。
We've observed incredibly strong results on production traffic across thousands of enterprise developers. During our early access period with dozens of enterprises, customers got frontier performance at approximately 30–50% lower cost.我们在数千名企业开发者及实际生产流量中观察到了极其出色的效果。在为期数十家企业的抢先体验阶段，客户以约 30%–50% 的成本降幅获得了顶尖的模型性能。
In online A/B tests across millions of requests, Cursor Router delivered frontier-quality performance at 60% savings.在跨越数百万次请求的在线 A/B 测试中，Cursor Router 在实现 60% 成本节省的同时，保持了顶尖的性能表现。
Cursor routes hundreds of millions of coding requests each week across every model and provider, with unique visibility into what users like and what stays in the codebase. Model neutrality has always been core to how Cursor works, and today we're putting that data and expertise to work for your team.Cursor 每周处理数以亿计的编码请求，涵盖了所有模型和提供商，因此我们对用户偏好以及代码库中的内容拥有独特的洞察力。模型中立性始终是 Cursor 运作的核心，今天，我们正将这些数据和专业经验转化为您团队的生产力。
With Cursor Router, our goal is to provide teams with the best performance and experience for every task, without spending more than the work requires.通过 Cursor Router，我们的目标是为团队提供每项任务的最佳性能和体验，同时确保支出不超出工作实际所需。
How it works# 工作原理
Roughly 60% of developers using Cursor pick a single model as their daily driver. This results in routine work being completed at frontier prices, and AI spend growing much faster than output quality. Cursor Router fixes that by classifying each request before a model runs.约 60% 的 Cursor 用户习惯将单一模型作为日常主力。这导致常规工作往往以顶尖模型的高昂价格完成，AI 支出增长速度远超产出质量的提升速度。Cursor Router 通过在模型运行前对每个请求进行分类，解决了这一问题。
At its core, Cursor Router is a classifier that routes users to the best model option based on their query. We trained Cursor Router on 600k+ live requests and evaluated performance in an online A/B test across millions of live requests directed by Cursor Router, optimizing for user satisfaction (AFC) as a reward.Cursor Router 的核心是一个分类器，它根据用户的查询将请求路由至最佳模型选项。我们使用超过 60 万个实时请求对 Cursor Router 进行了训练，并通过 Cursor Router 指导的数百万次实时请求进行了在线 A/B 测试，以用户满意度 (AFC) 作为奖励指标来优化性能。
Cursor Router analyzes each request on query, context, task complexity, and domain, combined with what we know about each model's behavior. We learn what each model is best at, and route to the most effective option. Simple work goes to the most price-efficient models, UI updates go to the model with the best taste, and more complex, long-horizon problems go to frontier reasoning models.Cursor Router 会结合查询内容、上下文、任务复杂度和领域，以及我们对各模型行为的了解，对每个请求进行分析。我们学习每个模型最擅长的领域，并将其路由至最有效的选项。简单的任务交给性价比最高的模型，UI 更新交给审美最好的模型，而更复杂、长周期的任务则交给具备顶尖推理能力的模型。
We designed our routing classifier for a world in which updated models get shipped early and often. This way as newer and more powerful models are released, we can easily update Cursor Router, so the experience keeps improving.我们设计的路由分类器旨在适应新模型频繁发布的世界。这样，随着更新、更强大的模型发布，我们可以轻松更新 Cursor Router，从而持续提升用户体验。
Cursor Router is cache-aware in both how it is trained and evaluated. It is trained on a dataset where routing results in cache misses, and evaluated in production where our reported cost savings include the cost of cache misses in routing decisions.Cursor Router 在训练和评估过程中都具备缓存感知能力。它是在路由结果导致缓存未命中的数据集上进行训练的，并在生产环境中进行评估，因此我们报告的成本节省已涵盖了路由决策中缓存未命中的成本。
Frontier intelligence at lower cost# 以更低成本实现顶尖智能
Cursor Router has three modes: Intelligence, Balance, and Cost which let you adjust where you are on the cost-intelligence Pareto frontier.Cursor Router 提供三种模式：Intelligence（智能）、Balance（平衡）和 Cost（成本），让您可以根据需求在成本与智能的帕累托前沿（Pareto frontier）上进行调整。
We found that Auto Intelligence mode lands near Fable on user satisfaction of output at about 60% lower cost for teams, while also lifting satisfaction about 15% over Opus 4.8 at nearly the same cost.我们发现，Auto Intelligence 模式在输出的用户满意度上接近 Fable 模型，但成本降低了约 60%；同时，与 Opus 4.8 相比，其满意度提升了约 15%，而成本几乎持平。
Similarly, Auto Balance lands above Opus 4.8 on user satisfaction with the results at about 36% lower cost. Against GPT-5.6 Sol, Auto Balance delivers comparable satisfaction at a lower spend rate.同样，Auto Balance 模式在用户满意度上超过了 Opus 4.8，且成本降低了约 36%。与 GPT-5.6 Sol 相比，Auto Balance 在支出更低的情况下提供了相当的满意度。
We chose to measure the efficacy of our router using large online A/B tests instead of offline evals. While offline evals are useful proxies for quality, they're limited by their small size, their distance from real-world usage, and the difficulty of reducing success to a rubric.我们选择通过大规模在线 A/B 测试而非离线评估来衡量路由器的效能。虽然离线评估是衡量质量的有用指标，但它们受限于规模小、与实际使用场景脱节以及难以将成功量化为评分标准等问题。
Offline evals also omit the extra cache-miss cost that comes from switching models. Real routing happens across a conversation: which model to pick, and when to switch.离线评估还忽略了切换模型时带来的额外缓存未命中成本。真实的路由发生在整个对话过程中：何时选择模型，以及何时进行切换。
Online A/B tests put Cursor Router to test in the real world across millions of tasks and conversations. Engineers write code, ask follow-ups, hit errors, and keep going, often across hundreds of requests in a week. Those are the conditions under which a model router needs to perform well.在线 A/B 测试将 Cursor Router 置于真实的数百万次任务和对话场景中进行检验。工程师在编写代码、提出后续问题、处理错误并持续推进工作的过程中，往往在一周内产生数百次请求。这些正是模型路由器需要表现出色的实际条件。
In terms of quality of output, we measured:在产出质量方面，我们衡量了以下指标：
- User satisfaction, classifying agent success based on user responses. Moving on to the next feature is a strong positive signal, while correcting the agent is a strong negative one.用户满意度：根据用户反馈对代理（Agent）的成功与否进行分类。继续进行下一个功能开发是强正向信号，而修正代理的行为则是强负向信号。
- Keep rate, or how much of the agent-generated code remains in the codebase over time.保留率：即代理生成的代码在代码库中长期保留的比例。
We have relied on these metrics to evaluate every model launch and harness improvement in the past nine months.在过去九个月里，我们一直依赖这些指标来评估每一次模型发布和工作流改进。
What customers are seeing# 客户反馈
Over the past two weeks, Cursor Router has been in early access with a selection of enterprise customers. We compared what they actually paid against the same traffic, priced entirely at Opus 4.8 API rates.在过去两周内，Cursor Router 已向部分企业客户开放抢先体验。我们将他们的实际支出与完全按照 Opus 4.8 API 费率计算的相同流量成本进行了对比。
In early access, three high-volume accounts with thousands of users saved 30%–50% on Auto-routed requests versus routing everything to Opus 4.8, with no decrease in quality.在抢先体验期间，三个拥有数千名用户的高流量账户在自动路由请求上节省了 30%–50% 的成本，且质量未受任何影响，而若全部使用 Opus 4.8 则无此节省。
Cost per request is only half the story. Engineering leaders care whether those savings show up in real shipped work, so we looked at cost per commit, and the pattern held.单次请求成本只是故事的一半。工程负责人更关心这些节省是否体现在实际产出的工作中，因此我们考察了单次提交（commit）的成本，结果显示这一模式同样成立。
For a single commit, we observed Cursor Router had a lower cost per commit of $6.76 for Intelligence mode and $4.63 for Balance.对于单次提交，我们观察到 Cursor Router 的单次提交成本更低，Intelligence 模式为 6.76 美元，Balance 模式为 4.63 美元。
GPT-5.6 Sol matched the cost of Intelligence but had lower user satisfaction with the output. Meanwhile, Fable 5 and Opus 4.8 produced commits at a cost premium to Cursor Router at $12.69 and $7.34 respectively.GPT-5.6 Sol 的成本与 Intelligence 模式相当，但用户对输出的满意度较低。与此同时，Fable 5 和 Opus 4.8 的提交成本则分别为 12.69 美元和 7.34 美元，高于 Cursor Router。
That gap is the practical case for routing. Cursor Router keeps hard tasks on the most capable models and moves routine work off of frontier pricing.这种差距正是路由的实际意义所在。Cursor Router 让高难度任务保留在能力最强的模型上，同时将常规工作从顶尖模型的昂贵定价中剥离出来。
You choose the tradeoff# 由您选择权衡方案
Cursor designed our router with teams and large organizations in mind. The router uses a data-driven taxonomy, while admins and end users can still choose where it sits on the cost-intelligence Pareto frontier.Cursor 在设计路由器时充分考虑了团队和大型组织的需求。该路由器使用数据驱动的分类法，同时管理员和最终用户仍可选择其在成本与智能帕累托前沿上的位置。
Select Auto mode in the model picker, and choose from three optimization modes that move you along the frontier:在模型选择器中选择 Auto 模式，并从三种优化模式中进行选择，以沿着前沿调整您的偏好：
- Intelligence: Frontier quality, with performance matching the most expensive and powerful models that might be out of reach for daily use.Intelligence：顶尖质量，性能媲美最昂贵、最强大的模型，这些模型可能因成本原因不适合日常使用。
- Balance: Strong quality, with performance matching the frontier models that most people like to daily drive.Balance：高质量，性能媲美大多数人喜爱的顶尖主力模型。
- Cost: Good quality, reaching the highest available intelligence while optimizing token spend.Cost：良好的质量，在优化 Token 支出的同时达到可用的最高智能水平。
Admins can decide how Cursor Router rolls out across teams. You can enable it per team or group, choose which modes members can select, set the default, and allow or block specific models.管理员可以决定 Cursor Router 如何在团队中推广。您可以按团队或群组启用它，选择成员可选的模式，设置默认值，并允许或禁止特定模型。
What’s next# 未来展望
Cursor Router is one piece of how Cursor drives token efficiency. Choosing the right model only matters if the agent itself stays lean, so we keep cutting waste in the harness around it.Cursor Router 只是 Cursor 推动 Token 效率提升的一环。只有当代理本身保持精简时，选择合适的模型才有意义，因此我们一直在削减其周围工作流中的浪费。
Dynamic tool calling is another clear example where most native tool descriptions are no longer loaded into every prompt. The model looks them up the first time it needs them, following the same pattern we already use for MCPs. This keeps common tools like read and edit hot while less commonly used tools only enter the prompt when the agent actually calls them.动态工具调用是另一个明显的例子，大多数原生工具描述不再被加载到每个 Prompt 中。模型会在第一次需要时查找这些描述，遵循我们已用于 MCP 的相同模式。这使得读取和编辑等常用工具保持活跃，而较少使用的工具仅在代理实际调用时才会进入 Prompt。
Alongside Cursor Router, we keep raising the floor and the ceiling of the model pool: Grok 4.5 widens what Cursor Router can draw from on harder, higher-cost work. Composer keeps getting better on the everyday path, so lower cost turns stay close to frontier quality without paying frontier prices.除了 Cursor Router，我们还在不断提升模型池的下限和上限：Grok 4.5 扩大了 Cursor Router 在处理更难、更高成本工作时的选择范围。Composer 在日常任务路径上的表现也在不断提升，因此低成本操作也能在不支付顶尖价格的情况下，保持接近顶尖的质量。
Cursor Router is available today for Teams and Enterprise plans across desktop, web, iOS, CLI, and our SDK.Cursor Router 即日起面向 Teams 和 Enterprise 计划用户开放，支持桌面端、Web 端、iOS、CLI 和我们的 SDK。
