{
  "schema_version": "1.1",
  "id": "archive:https://cursor.com/blog/how-cursor-router-works",
  "slug": "how-cursor-router-works-1fhejk3",
  "url": "https://feed7.dev/p/how-cursor-router-works-1fhejk3",
  "title": "How Cursor Router chooses the right model for the task",
  "why_included": "Cursor Router learns task complexity and model fit from production behavior, showing why agent routing should include correction signals, cache costs, and per-task performance.",
  "summary": "Cursor routes each turn in two stages: Compass estimates complexity, then a production-derived taxonomy selects among eligible frontier models. **Auto Intelligence costs 68% less than Fable**, while **Auto Balance costs 41% less than Opus 4.8** and reports higher satisfaction.",
  "practical_implication": "Builders implementing model routing should learn from real task outcomes, including corrections and task progression, rather than rely only on benchmarks. Include token use, cache misses, model-switching costs, confidence thresholds, and an explicit per-turn budget.",
  "agent_context": "Cursor routes each turn in two stages: Compass estimates complexity, then a production-derived taxonomy selects among eligible frontier models. **Auto Intelligence costs 68% less than Fable**, while **Auto Balance costs 41% less than Opus 4.8** and reports higher satisfaction.\n\nBuilders implementing model routing should learn from real task outcomes, including corrections and task progression, rather than rely only on benchmarks. Include token use, cache misses, model-switching costs, confidence thresholds, and an explicit per-turn budget.\n\nCursor's satisfaction measure is inferred from subsequent user behavior, so it is a proxy rather than a direct quality score. The reported results come from Cursor traffic and may not transfer to another workload, user population, or pricing mix.",
  "source": {
    "name": "Cursor",
    "url": "https://cursor.com/blog/how-cursor-router-works",
    "published_at": "2026-08-06T12:00:00.000Z"
  },
  "source_class": "blog_post",
  "content_type": "Engineering Post",
  "layer": "agent",
  "domains": [
    "coding",
    "data"
  ],
  "topics": [
    "model-selection",
    "harness-engineering",
    "agent-reliability"
  ],
  "verification": {
    "status": "official_source",
    "label": "Official Source",
    "method": "source_feed",
    "verified_at": null
  },
  "uncertainty": [
    "Cursor's satisfaction measure is inferred from subsequent user behavior, so it is a proxy rather than a direct quality score. The reported results come from Cursor traffic and may not transfer to another workload, user population, or pricing mix."
  ],
  "connected_context": {
    "meaning": "This turns model routing into a per-turn, budget-aware learning problem grounded in production behavior, not a static benchmark lookup. It contrasts directly with task-consistent routing: Cursor accepts switching costs to match each turn, while acknowledging that inferred satisfaction and workload-specific economics limit how broadly its reported savings can be applied.",
    "corpus_size": 390,
    "generated_at": "2026-08-08T10:05:17.014Z",
    "connections": [
      {
        "title": "TRACE-ROUTER: Task-Consistent and Adaptive Online Routing for Agentic AI",
        "source_name": "arXiv",
        "source_url": "https://arxiv.org/abs/2607.22465v1",
        "feed7_url": "https://feed7.dev/p/2607-22465v1-1g7nw7j",
        "reason": "TRACE-Router keeps one backend for an entire task, directly contrasting with Cursor’s per-turn selection and making model-switching cost versus finer-grained adaptation the central design choice."
      },
      {
        "title": "TurnSight: Turn-Level Hindsight Self-Distillation for Tool-Integrated Reasoning",
        "source_name": "arXiv",
        "source_url": "https://arxiv.org/abs/2608.04007v1",
        "feed7_url": "https://feed7.dev/p/2608-04007v1-0gjxuy2",
        "reason": "TurnSight reinforces the value of turn-level signals, but uses execution hindsight for training while Cursor infers routing quality from subsequent user behavior."
      },
      {
        "title": "Data and Environment Curation for Post-Training LLMs — Mahesh Sathiamoorthy, Bespoke Labs",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=ewtOo0scUh0",
        "feed7_url": "https://feed7.dev/p/data-and-environment-curation-for-post-training-llms-mahesh-sathiamoorth-19ab77a",
        "reason": "Both argue that improvement should come from real task outcomes and carefully designed data signals rather than benchmark performance alone; the post-training work also underscores the need to check whether learned routing transfers across distributions."
      }
    ]
  },
  "lifecycle": "Current",
  "published_at": "2026-08-06T12:00:00.000Z",
  "modified_at": "2026-08-06T12:00:00.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/how-cursor-router-works-1fhejk3",
    "json": "https://feed7.dev/p/how-cursor-router-works-1fhejk3.json",
    "markdown": "https://feed7.dev/p/how-cursor-router-works-1fhejk3.md"
  }
}