{
  "abstract": "Agentic systems increasingly delegate model selection to a router, yet open-source routers are usually evaluated with different tasks, candidate pools, and execution protocols, limiting direct comparison. We present a common measurement protocol and hybrid evaluation of four router implementations across RouterBench, BFCL v4, tau2-bench, and WebArena. We evaluate 290 frozen tasks against a locked matrix of 2,610 candidate outcomes. Three routers emit constant or near-constant tier assignments; only vLLM Semantic Router varies materially with prompt content, and it has the highest observed success rate on none of the four benchmarks. Always-Mid matches Aurelio exactly on three benchmarks and within 0.003 on the fourth. For vLLM, task-level superiority tests detect no task-specific advantage over a share-matched content-blind allocation; equivalence is established only on WebArena at the protocol-declared five-percentage-point margin. The results show that, under these configurations and controls, observed gains track selected-tier composition more closely than demonstrated task-specific targeting. Fixed-tier baselines and selected-tier distributions are therefore necessary controls in router evaluation; the findings are scoped to these configurations, candidate pool, and frozen benchmark samples, not to routing paradigms in general.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2608.14641",
  "author_total": 2,
  "authors": [
    {
      "name": "Kiran N. Kumar"
    },
    {
      "name": "Santhosh K. Saminathan"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/4ahgff7ameaft.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2608.14641",
  "fulltext_url": null,
  "id": "4ahgff7ameaft",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2608.14641",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.AI"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "openalex",
      "1702|Topic Modeling"
    ]
  ],
  "openalex": "W7203699662",
  "pdf_url": "https://arxiv.org/pdf/2608.14641",
  "pmcid": null,
  "pmid": null,
  "published": "2026-08-01",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/4ahgff7ameaft.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:20:36Z",
      "native_id": "2608.14641",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2608.14641"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "Task- and Session-Level Model Routing: A Common-Interface Hybrid Evaluation of Four Open-Source Routers Across Four Benchmarks",
  "title_text": "Task- and Session-Level Model Routing: A Common-Interface Hybrid Evaluation of Four Open-Source Routers Across Four Benchmarks",
  "type": "preprint",
  "updated_at": "2026-10-30T12:20:36Z",
  "url": "https://research.draconitic.com/paper/4ahgff7ameaft",
  "venue": "arXiv",
  "versions": []
}