{
  "schema_version": "newruntime-topic-hub-v0.1",
  "type": "topic_hub",
  "slug": "model-routing",
  "title": "Model routing - New Runtime",
  "description": "How teams route work across model families, effort levels, cheap executors, judges, and local/runtime constraints.",
  "answer": [
    "Model routing treats models as swappable execution resources rather than a single product choice.",
    "The pattern is to match task risk, latency, cost, and verification needs to the right model or model sequence.",
    "This makes the harness more durable than any one frontier model release."
  ],
  "search_intents": [
    "model routing",
    "multi model AI systems",
    "LLM routing architecture"
  ],
  "status": "featured",
  "last_updated": "2026-07-16",
  "counts": {
    "total": 13,
    "signals": 11,
    "patterns": 2,
    "posts": 0,
    "atlas": 0,
    "sources": 25
  },
  "routes": {
    "html": "https://newruntime.com/topics/model-routing/",
    "markdown": "https://newruntime.com/topics/model-routing.md",
    "json": "https://newruntime.com/topics/model-routing.json"
  },
  "top_sources": [
    "https://claude.com/blog/claude-model-and-effort-level-in-claude-code",
    "https://code.claude.com/docs/en/model-config",
    "https://cognition.com/blog/making-fable-cheaper-than-opus",
    "https://databricks.com/blog/benchmarking-coding-agents-databricks-multi-million-line-codebase",
    "https://docs.inference.net/introduction",
    "https://engineering.ramp.com/post/ai-spend-value",
    "https://github.com/langchain-ai/deepagents",
    "https://github.com/NousResearch/hermes-agent",
    "https://github.com/rivet-dev/agentos",
    "https://github.com/shadcn/improve",
    "https://github.com/shadcn/improve/blob/main/skills/improve/SKILL.md",
    "https://github.com/steipete/agent-scripts/blob/main/skills/codex-first/SKILL.md"
  ],
  "patterns": [
    {
      "kind": "Pattern",
      "slug": "harness-architecture-outlives-model-choice",
      "title": "Harness architecture outlives model choice",
      "description": "For production agents, the harness is becoming a more durable product boundary than the identity of the model running inside it.",
      "date": "2026-07-16",
      "topics": [
        "harness-engineering",
        "agent-runtime",
        "model-routing"
      ],
      "source_count": 5,
      "metric": "high",
      "url": "https://newruntime.com/patterns/harness-architecture-outlives-model-choice/"
    },
    {
      "kind": "Pattern",
      "slug": "agent-economics-moves-to-completed-work",
      "title": "Agent economics moves to completed work",
      "description": "The economically meaningful unit for agent systems is becoming cost per verified completed task rather than cost per token or model call.",
      "date": "2026-07-15",
      "topics": [
        "agent-economics",
        "model-routing",
        "product-metrics"
      ],
      "source_count": 5,
      "metric": "high",
      "url": "https://newruntime.com/patterns/agent-economics-moves-to-completed-work/"
    }
  ],
  "field_notes": [],
  "raw_signals": [
    {
      "kind": "Raw Signal",
      "slug": "opencode-model-swappable-agent-harness",
      "title": "OpenCode makes the model a swappable dependency",
      "description": "OpenCode keeps rules, skills, permissions, sessions, and tools stable while the underlying model changes.",
      "date": "2026-07-16",
      "topics": [
        "coding-agents",
        "harness-engineering",
        "model-routing",
        "developer-tools"
      ],
      "source_count": 4,
      "metric": "structural",
      "url": "https://newruntime.com/signals/opencode-model-swappable-agent-harness/"
    },
    {
      "kind": "Raw Signal",
      "slug": "fable-cost-per-completed-task",
      "title": "A pricier model can be cheaper per completed task",
      "description": "Cognition reports that Fable 5 completed coding work with fewer steps and output tokens than its previous lead model.",
      "date": "2026-07-15",
      "topics": [
        "agent-economics",
        "coding-agents",
        "model-routing"
      ],
      "source_count": 1,
      "metric": "structural",
      "url": "https://newruntime.com/signals/fable-cost-per-completed-task/"
    },
    {
      "kind": "Raw Signal",
      "slug": "claude-code-model-and-effort-are-separate-controls",
      "title": "Claude Code separates model choice from effort",
      "description": "Anthropic exposes model selection and effort level as different controls for capability, token use, latency, and persistence.",
      "date": "2026-07-13",
      "topics": [
        "coding-agents",
        "model-routing",
        "agent-economics"
      ],
      "source_count": 2,
      "metric": "notable",
      "url": "https://newruntime.com/signals/claude-code-model-and-effort-are-separate-controls/"
    },
    {
      "kind": "Raw Signal",
      "slug": "improve-routes-architect-and-executor-models",
      "title": "Improve routes architecture and execution to different models",
      "description": "The Improve skill audits a repository with a stronger planning model and delegates bounded fixes to cheaper executors.",
      "date": "2026-07-12",
      "topics": [
        "agent-skills",
        "model-routing",
        "code-review"
      ],
      "source_count": 2,
      "metric": "notable",
      "url": "https://newruntime.com/signals/improve-routes-architect-and-executor-models/"
    },
    {
      "kind": "Raw Signal",
      "slug": "advisor-model-guides-cheaper-executor",
      "title": "An advisor model can guide a cheaper executor",
      "description": "The advisor-tool pattern lets a fast executor request bounded analysis from a stronger model while keeping control of the task loop.",
      "date": "2026-07-09",
      "topics": [
        "model-routing",
        "orchestration",
        "agent-economics"
      ],
      "source_count": 2,
      "metric": "notable",
      "url": "https://newruntime.com/signals/advisor-model-guides-cheaper-executor/"
    },
    {
      "kind": "Raw Signal",
      "slug": "inference-gateway-makes-model-switching-operational",
      "title": "An inference gateway makes model switching operational",
      "description": "Inference.net exposes model routing through a stable gateway so clients can change providers without rewriting every integration.",
      "date": "2026-07-02",
      "topics": [
        "model-routing",
        "inference",
        "agent-infrastructure"
      ],
      "source_count": 1,
      "metric": "notable",
      "url": "https://newruntime.com/signals/inference-gateway-makes-model-switching-operational/"
    },
    {
      "kind": "Raw Signal",
      "slug": "coinbase-reports-lower-ai-spend-as-token-use-rises",
      "title": "Coinbase Reports Lower AI Spend as Token Use Rises",
      "description": "A production account suggests routing, caching, model choice, and workflow design can reduce unit cost even while aggregate agent usage expands.",
      "date": "2026-06-28",
      "topics": [
        "ai-economics",
        "model-routing",
        "enterprise-ai"
      ],
      "source_count": 1,
      "metric": "notable",
      "url": "https://newruntime.com/signals/coinbase-reports-lower-ai-spend-as-token-use-rises/"
    },
    {
      "kind": "Raw Signal",
      "slug": "glm-5-2-changes-the-economics-of-claude-code-workflows",
      "title": "GLM-5.2 Changes the Economics of Claude Code Workflows",
      "description": "A capable open model routed through compatible infrastructure can replace premium inference for parts of a coding-agent workload.",
      "date": "2026-06-27",
      "topics": [
        "open-models",
        "model-routing",
        "coding-agents"
      ],
      "source_count": 4,
      "metric": "notable",
      "url": "https://newruntime.com/signals/glm-5-2-changes-the-economics-of-claude-code-workflows/"
    },
    {
      "kind": "Raw Signal",
      "slug": "sakana-fugu-presents-multiple-agents-as-one-model-endpoint",
      "title": "Sakana Fugu Presents Multiple Agents as One Model Endpoint",
      "description": "Fugu hides a multi-agent debate and synthesis system behind a model-like interface, making orchestration an implementation detail of inference.",
      "date": "2026-06-25",
      "topics": [
        "multi-agent",
        "model-routing",
        "inference"
      ],
      "source_count": 1,
      "metric": "notable",
      "url": "https://newruntime.com/signals/sakana-fugu-presents-multiple-agents-as-one-model-endpoint/"
    },
    {
      "kind": "Raw Signal",
      "slug": "fusion-api-uses-multiple-models-and-a-judge-for-one-answer",
      "title": "Fusion API Uses Multiple Models and a Judge for One Answer",
      "description": "A single endpoint can route a question to several models and synthesize their outputs, moving model selection and comparison behind an orchestration layer.",
      "date": "2026-06-17",
      "topics": [
        "model-routing",
        "multi-model",
        "evaluation"
      ],
      "source_count": 1,
      "metric": "notable",
      "url": "https://newruntime.com/signals/fusion-api-uses-multiple-models-and-a-judge-for-one-answer/"
    },
    {
      "kind": "Raw Signal",
      "slug": "vercel-treats-production-ai-as-multi-model-infrastructure",
      "title": "Vercel Treats Production AI as Multi-Model Infrastructure",
      "description": "Vercel's gateway view assumes model providers are routable dependencies with fallback, observability, and policy around them.",
      "date": "2026-05-15",
      "topics": [
        "model-routing",
        "ai-gateway",
        "production-ai"
      ],
      "source_count": 1,
      "metric": "notable",
      "url": "https://newruntime.com/signals/vercel-treats-production-ai-as-multi-model-infrastructure/"
    }
  ],
  "atlas_records": []
}