Working with Multiple Agents

LangDB AI Gateway offers plug and play observability when you work with AI agents.

LangDB automatically visualizes how agents interact, providing a clear view of workflows, hierarchies, and usage patterns.

LangDB supports three main concepts.

  • Run: A complete end-to-end interaction between agents, grouped for easy tracking.

  • Thread: Aggregate multiple Runs into a single thread for a unified chat experience.

  • Parent Trace ID: Optionally link different traces to build hierarchical relationships and visualize dependencies.

Example

from openai import OpenAI
from uuid import uuid4
client = OpenAI(
    base_url="https://api.us-east-1.langdb.ai"  # LangDB API base URL,
    api_key=api_key,  # Replace with your LangDB token
)

response = client.chat.completions.create(
    model="gpt-4o",  # Use the model
    messages=[{"role": "developer", "content": "You are a helpful assistant."},
              {"role": "user", "content": "Hello!"}],  
    extra_headers={"x-project-id": "xxxxx"
                   "x-thread-id": str(uuid4()),
                   "x-run-id": str(uuid4())}
)

What You Can Do with LangDB

  • Monitor Interactions: Track how agents communicate and collaborate within your workflows.

  • Optimize System Performance: Detect inefficiencies or bottlenecks and refine your workflows for better outcomes.

  • Control Costs: Analyze token usage and associated costs to optimize expenditure across workflows.

  • Debug Seamlessly: Quickly pinpoint and resolve issues with trace visualizations.

Checkout the full Multi-Agent Tracing Example here.

Updated on