Switchcraft: AI Model Router for Agentic Tool Calling
cs.AI, cs.MA
Submitted: 2026-05-08
Updated: 2026-10-30
Code: https://github.com/vllm-project/semantic-router
Terminology
Sources
- RouteNator: A Router-Based Multi-Modal Architecture for Generating Synthetic Training Data for Function Calling LLMs
- IPR: Intelligent Prompt Routing with User-Controlled Quality-Cost Trade-offs
- GraphRouter: A Graph-based Router for LLM Selections
- SWE-bench: Can Language Models Resolve Real-World GitHub Issues?
- Universal Model Routing for Efficient LLM Inference
- LLMRouterBench: A Massive Benchmark and Unified Framework for LLM Routing
- RouterArena: An Open Platform for Comprehensive Comparison of LLM Routers
- OmniRouter: Budget and Performance Controllable Multi-LLM Routing
- MetaLLM: A High-performant and Cost-efficient Dynamic Framework for Wrapping LLMs
- Explainable Model Routing for Agentic Workflows
- DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter
- Large Language Model Routing with Benchmark Datasets
- CARROT: A Cost Aware Rate Optimal Router
- TensorOpera Router: A Multi-Model Router for Efficient LLM Inference
- R2-Router: A New Paradigm for LLM Routing with Reasoning
- $\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains
- xLAM: A Family of Large Action Models to Empower AI Agent Systems
- EmbedLLM: Learning Compact Representations of Large Language Models
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents