Tuning Deep Agents to Work Well with Different Models

About This Tutorial

LangChain Blog has introduced model-specific profiles for Deep Agents to improve performance with various AI models. The new profiles are designed to adjust prompts, tools, and middleware for OpenAI, Anthropic, and Google models. This update has led to a significant performance boost, with a 10-20 point jump on a subset of tau2-bench compared to the default harness. The change aims to enhance the flexibility and effectiveness of Deep Agents in different deployment scenarios. The update is now available out of the box, allowing users to take advantage of the improved performance...