Awesome-Routing-LLMs

A curated list of awesome works in Routing LLMs paradigm

🤩👉 Welcome to submit your contributions to this code repository.

Routing LLM Paper (preprint)

Paper	Conference/Journal	Code	Type
Dynamic Model Routing and Cascading for Efficient LLM Inference: A Survey	arxiv'26	-	Survey
LLMRouterBench: A Massive Benchmark and Unified Framework for LLM Routing	arxiv'26	Link	Benchmark
Arch-Router: Aligning LLM Routing with Human Preferences	arxiv'25	Link	Method
RadialRouter: Structured Representation for Efficient and Robust Large Language Models Routing	arxiv'25	-	Method
LLM-Router: Learning to Route LLMs with Confidence Tokens	arxiv'25	-	Method
OmniRouter: Budget and Performance Controllable Multi-LLM Routing	arxiv'25	Link	Method
CP-Router: An Uncertainty-Aware Router Between LLM and LRM	arxiv'25	-	Method
Confident or Seek Stronger: Exploring Uncertainty-Based On-device LLM Routing From Benchmarking to Generalization	arxiv'25	-	Method
Route-and-Reason: Scaling Large Language Model Reasoning with Reinforced Model Router	arxiv'25	Link	Method
LightRouter: Towards Efficient LLM Collaboration with Minimal Overhead	arxiv'25	-	Method
Universal Model Routing for Efficient LLM Inference	arxiv'25	-	Method
MasRouter: Learning to Route LLMs for Multi-Agent Systems	arxiv'25	Link	Method
Leveraging Uncertainty Estimation for Efficient LLM Routing	arxiv'25	-	Method
Prompt-to-Leaderboard	arxiv'25	Link	Method
MixLLM: Dynamic Routing in Mixed Large Language Models	arxiv'25	-	Method
MetaLLM: A High-performant and Cost-efficient Dynamic Framework for Wrapping LLMs	arxiv'24	-	Method
Merge, Ensemble, and Cooperate! A Survey on Collaborative Strategies in the Era of Large Language Models	arxiv'24	-	Survey
Routoo: Learning to Route to Large Language Models Effectively	arxiv'24	-	Method
Tryage: Real-time, intelligent Routing of User Prompts to Large Language Models	arxiv'23	-	Method
FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance	arxiv'23	Link	Method
vLLM Semantic Router	Website	Link	Method
Aurelio Semantic Router	Website	Link	Method

Routing LLM Paper

Paper	Conference/Journal	Code	Type
ICL-Router: In-Context Learned Model Representations for LLM Routing	AAAI'26	Link	Method
The Avengers: A Routing Recipe for Collective Intelligence in Language Models	AAAI'26	Link	Method
Beyond GPT-5: Making LLMs Cheaper and Better via Performance–Efficiency Optimized Routing	DAI'25	Link	Method
MESS+: Dynamically Learned Inference-Time LLM Routing in Model Zoos with Service Level Guarantees	NeurIPS'25	Link	Method
Router-R1: Teaching LLMs Multi-Round Routing and Aggregation via Reinforcement Learning	NeurIPS'25	Link	Method
Adaptive LLM Routing Under Budget Constraints	EMNLP'25	-	Method
RouterEval: A Comprehensive Benchmark for Routing LLMs to Explore Model-level Scaling Up in LLMs	EMNLP'25	Link	Benchmark
IRT-Router: Effective and Interpretable Multi-LLM Routing via Item Response Theory	ACL'25	Link	Method
TagRouter: Learning Route to LLMs through Tags for Open-Domain Text Generation Tasks	ACL'25	-	Method
GraphRouter: A Graph-based Router for LLM Selections	ICLR'25	Link	Method
BEST-Route: Adaptive LLM Routing with Test-Time Optimal Compute	ICML'25	-	Method
Capability Instruction Tuning: A New Paradigm for Dynamic LLM Routing	AAAI'25	Link	Method
Universal LLM Routing with Correctness-Based Representation	ICLR'25	-	Method
EmbedLLM: Learning Compact Representations of Large Language Models	ICLR'25	Link	Method
RouteLLM: Learning to Route LLMs from Preference Data	ICLR'25	Link	Method
TensorOpera Router: A Multi-Model Router for Efficient LLM Inference	EMNLP'24	-	Method
Cache & Distil: Optimising API Calls to Large Language Models	ACL'24	-	Method
Model-GLUE: Democratized LLM Scaling for A Large Model Zoo in the Wild	NeurIPS'24	Link	Method
RouterDC: Query-Based Router by Dual Contrastive Learning for Assembling Large Language Models	NeurIPS'24	Link	Method
AutoMix: Automatically Mixing Language Models	NeurIPS'24	Link	Method
Routing to the Expert: Efficient Reward-guided Ensemble of Large Language Models	NAACL'24	-	Method
Harnessing the Power of Multiple Minds: Lessons Learned from LLM Routing	NAACLW'24	Link	Method
Large Language Model Routing with Benchmark Datasets	COLM'24	-	Method
Optimising Calls to Large Language Models with Uncertainty-Based Two-Tier Selection	COLM'24	Link	Method
OptLLM: Optimal Assignment of Queries to Large Language Models	ICWS'24	Link	Method
RouterBench: A Benchmark for Multi-LLM Routing System	ICMLW'24	Link	Benchmark
Fly-Swat or Cannon? Cost-Effective Language Model Choice via Meta-Modeling	WSDM'24	Link	Method
Hybrid LLM: Cost-Efficient and Quality-Aware Query Routing	ICLR'24	-	Method
Model Spider: Learning to Rank Pre-Trained Models Efficiently	NeurIPS'23	Link	Method
LLM-Blender: Ensembling Large Language Models with Pairwise Ranking and Generative Fusion	ACL'23	Link	Method