{"id":"https://openalex.org/W7203699662","doi":"https://doi.org/10.48550/arxiv.2608.14641","title":"Task- and Session-Level Model Routing: A Common-Interface Hybrid Evaluation of Four Open-Source Routers Across Four Benchmarks","display_name":"Task- and Session-Level Model Routing: A Common-Interface Hybrid Evaluation of Four Open-Source Routers Across Four Benchmarks","publication_year":2026,"publication_date":"2026-07-28","ids":{"openalex":"https://openalex.org/W7203699662","doi":"https://doi.org/10.48550/arxiv.2608.14641"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.14641","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.14641","pdf_url":"https://arxiv.org/pdf/2608.14641","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.14641","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5147400179","display_name":"Kiran N. Kumar","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210119109","display_name":"Indiana University Bloomington","ror":"https://ror.org/02k40bc56","country_code":"US","type":"education","lineage":["https://openalex.org/I4210119109","https://openalex.org/I592451"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kumar, Kiran N.","raw_affiliation_strings":["Indiana University , Bloomington , IN , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Indiana University , Bloomington , IN , USA","institution_ids":["https://openalex.org/I4210119109"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5147391271","display_name":"Santhosh K. Saminathan","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Saminathan, Santhosh K.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.3093000054359436,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.3093000054359436,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10286","display_name":"Information Retrieval and Search Behavior","score":0.15029999613761902,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.1006999984383583,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/model-routing","display_name":"model routing","score":0.6100000143051147},{"id":"https://openalex.org/keywords/agentic-systems","display_name":"agentic systems","score":0.6240000128746033},{"id":"https://openalex.org/keywords/model-selection","display_name":"model selection","score":0.5690000057220459},{"id":"https://openalex.org/keywords/webarena","display_name":"WebArena","score":0.9959999918937683}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7598999738693237},{"id":"https://openalex.org/C2775896111","wikidata":"https://www.wikidata.org/wiki/Q642560","display_name":"Router","level":2,"score":0.6862999796867371},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5685999989509583},{"id":"https://openalex.org/C143273055","wikidata":"https://www.wikidata.org/wiki/Q2382794","display_name":"Delegate","level":2,"score":0.5153999924659729},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.47209998965263367},{"id":"https://openalex.org/C2780385302","wikidata":"https://www.wikidata.org/wiki/Q367158","display_name":"Protocol (science)","level":3,"score":0.42719998955726624},{"id":"https://openalex.org/C184896649","wikidata":"https://www.wikidata.org/wiki/Q290066","display_name":"Routing table","level":4,"score":0.4194999933242798},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.4059999883174896},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.39820000529289246},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.3930000066757202},{"id":"https://openalex.org/C104954878","wikidata":"https://www.wikidata.org/wiki/Q1648707","display_name":"Routing protocol","level":3,"score":0.38530001044273376},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.383899986743927},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.3734999895095825},{"id":"https://openalex.org/C38601921","wikidata":"https://www.wikidata.org/wiki/Q1757693","display_name":"Protocol stack","level":3,"score":0.3614000082015991},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.35850000381469727},{"id":"https://openalex.org/C2780069185","wikidata":"https://www.wikidata.org/wiki/Q7977945","display_name":"Equivalence (formal languages)","level":2,"score":0.30489999055862427},{"id":"https://openalex.org/C43711488","wikidata":"https://www.wikidata.org/wiki/Q7534783","display_name":"Skew","level":2,"score":0.29109999537467957},{"id":"https://openalex.org/C175291020","wikidata":"https://www.wikidata.org/wiki/Q1156822","display_name":"Offset (computer science)","level":2,"score":0.28859999775886536},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2865999937057495},{"id":"https://openalex.org/C45235069","wikidata":"https://www.wikidata.org/wiki/Q278425","display_name":"Table (database)","level":2,"score":0.28299999237060547},{"id":"https://openalex.org/C25343380","wikidata":"https://www.wikidata.org/wiki/Q277521","display_name":"Relation (database)","level":2,"score":0.2824999988079071},{"id":"https://openalex.org/C31395832","wikidata":"https://www.wikidata.org/wiki/Q1318674","display_name":"Testbed","level":2,"score":0.2786000072956085},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.26429998874664307},{"id":"https://openalex.org/C178086212","wikidata":"https://www.wikidata.org/wiki/Q11155","display_name":"Border Gateway Protocol","level":5,"score":0.2535000145435333}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.14641","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.14641","pdf_url":"https://arxiv.org/pdf/2608.14641","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.14641","doi":"https://doi.org/10.48550/arxiv.2608.14641","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.14641","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.14641","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.14641","pdf_url":"https://arxiv.org/pdf/2608.14641","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7203699662.pdf","grobid_xml":"https://content.openalex.org/works/W7203699662.grobid-xml"},"referenced_works_count":11,"referenced_works":["https://openalex.org/W4376122773","https://openalex.org/W4387163503","https://openalex.org/W4392971790","https://openalex.org/W4399836817","https://openalex.org/W4400141368","https://openalex.org/W4415795271","https://openalex.org/W4417299034","https://openalex.org/W7123470302","https://openalex.org/W7128633360","https://openalex.org/W7151448928","https://openalex.org/W7166872418"],"related_works":["https://openalex.org/W7161871162","https://openalex.org/W4417299034","https://openalex.org/W4414804952","https://openalex.org/W7140114980","https://openalex.org/W7128772504","https://openalex.org/W7223144003","https://openalex.org/W7204176819","https://openalex.org/W7124118279","https://openalex.org/W7207775301","https://openalex.org/W4392971790"],"abstract_inverted_index":{"Agentic":[0],"systems":[1],"increasingly":[2],"delegate":[3],"model":[4],"selection":[5],"to":[6,164,174],"a":[7,28,52,113],"router,":[8],"yet":[9],"open-source":[10],"routers":[11,60],"are":[12,153,162],"usually":[13],"evaluated":[14],"with":[15,73],"different":[16],"tasks,":[17],"candidate":[18,57,167],"pools,":[19],"and":[20,32,44,76,97,135,150,169],"execution":[21],"protocols,":[22],"limiting":[23],"direct":[24],"comparison.":[25],"We":[26,46],"present":[27],"common":[29],"measurement":[30],"protocol":[31],"hybrid":[33],"evaluation":[34],"of":[35,55,86],"four":[36,88],"router":[37,158],"implementations":[38],"across":[39],"RouterBench,":[40],"BFCL":[41],"v4,":[42],"tau2-bench,":[43],"WebArena.":[45],"evaluate":[47],"290":[48],"frozen":[49,170],"tasks":[50],"against":[51],"locked":[53],"matrix":[54],"2,610":[56],"outcomes.":[58],"Three":[59],"emit":[61],"constant":[62],"or":[63],"near-constant":[64],"tier":[65],"assignments;":[66],"only":[67,120],"vLLM":[68],"Semantic":[69],"Router":[70],"varies":[71],"materially":[72],"prompt":[74],"content,":[75],"it":[77],"has":[78],"the":[79,87,101,124,160],"highest":[80],"observed":[81,137],"success":[82],"rate":[83],"on":[84,94,100,121],"none":[85],"benchmarks.":[89],"Always-Mid":[90],"matches":[91],"Aurelio":[92],"exactly":[93],"three":[95],"benchmarks":[96],"within":[98],"0.003":[99],"fourth.":[102],"For":[103],"vLLM,":[104],"task-level":[105],"superiority":[106],"tests":[107],"detect":[108],"no":[109],"task-specific":[110,146],"advantage":[111],"over":[112],"share-matched":[114],"content-blind":[115],"allocation;":[116],"equivalence":[117],"is":[118],"established":[119],"WebArena":[122],"at":[123],"protocol-declared":[125],"five-percentage-point":[126],"margin.":[127],"The":[128],"results":[129],"show":[130],"that,":[131],"under":[132],"these":[133,165],"configurations":[134],"controls,":[136],"gains":[138],"track":[139],"selected-tier":[140,151],"composition":[141],"more":[142],"closely":[143],"than":[144],"demonstrated":[145],"targeting.":[147],"Fixed-tier":[148],"baselines":[149],"distributions":[152],"therefore":[154],"necessary":[155],"controls":[156],"in":[157,177],"evaluation;":[159],"findings":[161],"scoped":[163],"configurations,":[166],"pool,":[168],"benchmark":[171],"samples,":[172],"not":[173],"routing":[175],"paradigms":[176],"general.":[178]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-08-19T00:00:00"}