agent-routing

LangChain shares a practical guide on building a model router inside the agent harness—covering task understanding, model selection, and outcome tracking—claiming a 64% reduction in median cost per thread with no quality loss.

CData has launched an AI gateway for agents that routes requests, enforces policies, and maintains a context graph to preserve enterprise knowledge across model changes.

Cua has open-sourced CUA-S1-FORMS, a tiny 2.8MB AI model specialized for form-filling tasks, achieving 99.7% accuracy and enabling local deployment.

A developer asks for architectural recommendations on building a multi-agent LangGraph application with a supervisor to handle dynamic routing, state management, and human-in-the-loop workflows within the same chat session.

A developer describes building a multi-engine agentic coding harness that routes sessions across Claude Code, Codex, and Ollama, selecting from twelve models per session based on task value.

Multi-agent AI systems commonly fail at routing, parallelism, handoffs, and coverage. This post recommends a dispatch matrix, parallel execution, structured handoffs, and a catch-all fallback with logging to fix these issues.

The Kimi K2.7 Code High Speed model offers 5x throughput at 2x cost, leading to selective routing within an agent system.

This paper studies how routing accuracy degrades as the number of agents scales from 10 to 110 in an enterprise productivity assistant, finding F1 drops of 16–23 percentage points. It diagnoses retrieval and confusion gaps and shows that embedding-based shortlisting recovers 10–11pp F1.

Introduces OpenClaude, an open-source alternative that supports multiple models like DeepSeek, GPT-4, etc., and uses Agent Routing to intelligently assign tasks, saving money and improving efficiency.

This paper introduces BoundaryRouter, a training-free framework that optimizes LLM agent usage by routing queries to either lightweight inference or full agent execution based on early experience. It also presents RouteBench, a benchmark for evaluating routing performance, showing significant improvements in speed and accuracy.