@inproceedings{jiangeng-etal-2026-systematic,
title = "A Systematic Survey of Large Language Model Routing",
author = "Jiangeng, Chang and
Wu, Xingyu and
Tan, KC and
Feng, Mengling",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.2072/",
pages = "44102--44134",
ISBN = "979-8-89176-462-0",
abstract = "Available LLMs differ substantially in cost, latency, and capability. Their strengths vary across queries, so the natural question becomes: which model for which query? This is the LLM routing problem. Routing is already used in production, and the academic literature has grown across cascading, bandit, generative, and benchmark threads, each with its own vocabulary, each largely ignoring the others. We argue the field is best read along a single axis: the \textit{algorithmic operation} the router performs. We organise the recurring operations under four labels. Predictive routers score candidates before their answers are generated. Adaptive routers try a model, observe the outcome, and adjust. Generative routers let an LLM itself reason about the decision. Infrastructure work studies routing rather than performs it, via benchmarks and adversarial analysis. This survey formalises the routing decision under this four-paradigm lens, traces the methodological development of the field, compares assumptions and tradeoffs across paradigms, and identifies open research directions revealed by the taxonomy. A continuously updated reading list lives at \url{https://github.com/wuxingyu-ai/LLM-Routing}."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="jiangeng-etal-2026-systematic">
<titleInfo>
<title>A Systematic Survey of Large Language Model Routing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chang</namePart>
<namePart type="family">Jiangeng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xingyu</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">K</namePart>
<namePart type="given">C</namePart>
<namePart type="family">Tan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mengling</namePart>
<namePart type="family">Feng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Available LLMs differ substantially in cost, latency, and capability. Their strengths vary across queries, so the natural question becomes: which model for which query? This is the LLM routing problem. Routing is already used in production, and the academic literature has grown across cascading, bandit, generative, and benchmark threads, each with its own vocabulary, each largely ignoring the others. We argue the field is best read along a single axis: the algorithmic operation the router performs. We organise the recurring operations under four labels. Predictive routers score candidates before their answers are generated. Adaptive routers try a model, observe the outcome, and adjust. Generative routers let an LLM itself reason about the decision. Infrastructure work studies routing rather than performs it, via benchmarks and adversarial analysis. This survey formalises the routing decision under this four-paradigm lens, traces the methodological development of the field, compares assumptions and tradeoffs across paradigms, and identifies open research directions revealed by the taxonomy. A continuously updated reading list lives at https://github.com/wuxingyu-ai/LLM-Routing.</abstract>
<identifier type="citekey">jiangeng-etal-2026-systematic</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.2072/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>44102</start>
<end>44134</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Systematic Survey of Large Language Model Routing
%A Jiangeng, Chang
%A Wu, Xingyu
%A Tan, K. C.
%A Feng, Mengling
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F jiangeng-etal-2026-systematic
%X Available LLMs differ substantially in cost, latency, and capability. Their strengths vary across queries, so the natural question becomes: which model for which query? This is the LLM routing problem. Routing is already used in production, and the academic literature has grown across cascading, bandit, generative, and benchmark threads, each with its own vocabulary, each largely ignoring the others. We argue the field is best read along a single axis: the algorithmic operation the router performs. We organise the recurring operations under four labels. Predictive routers score candidates before their answers are generated. Adaptive routers try a model, observe the outcome, and adjust. Generative routers let an LLM itself reason about the decision. Infrastructure work studies routing rather than performs it, via benchmarks and adversarial analysis. This survey formalises the routing decision under this four-paradigm lens, traces the methodological development of the field, compares assumptions and tradeoffs across paradigms, and identifies open research directions revealed by the taxonomy. A continuously updated reading list lives at https://github.com/wuxingyu-ai/LLM-Routing.
%U https://aclanthology.org/2026.findings-emnlp.2072/
%P 44102-44134
Markdown (Informal)
[A Systematic Survey of Large Language Model Routing](https://aclanthology.org/2026.findings-emnlp.2072/) (Jiangeng et al., Findings 2026)
ACL
- Chang Jiangeng, Xingyu Wu, KC Tan, and Mengling Feng. 2026. A Systematic Survey of Large Language Model Routing. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 44102–44134, Budapest, Hungary. Association for Computational Linguistics.