DOCSEARCHbyAlgolia
Submit Docs

LangChain Overview

website·Indexed 18 days ago

https://js.langchain.com/v0.1/docs/get_started/introduction/

Documentation providing an overview of LangChain, including core benefits and instructions for creating agents.

Tokens
846
Snippets
3
Records
8
Agent score
49%

What's inside LangChain

  4. Choose between LangChain, LangGraph, and Deep Agents

    v0.1

    Depending on the level of control and automation needed, choose the appropriate framework:

    • Deep Agents: Use for 'batteries-included' agents with built-in context compression, virtual filesystems, and subagent-spawning.
    • LangChain (create_agent): Use for a highly customizable agent harness that can be tailored to specific use cases and data.
    • LangGraph: Use for low-level orchestration of advanced workflows combining deterministic and agentic logic.
    • LangSmith: Use across any of these frameworks for tracing, debugging, and evaluating agent behavior.

  7. Create a custom agent using createAgent

    v0.1
    LangChain provides the createAgent function to build a minimal, configurable agent harness. An agent is composed of a model, tools, a prompt, and middleware. You can define custom tools using the tool function and a Zod schema for input validation.
    // Example using OpenAI
// Install: npm install langchain zod @langchain/openai
import { createAgent, tool } from "langchain";
import * as z from "zod";

const getWeather = tool(
  (input) => `It's always sunny in ${input.city}!`,
  {
    name: "get_weather",
    description: "Get the weather for a given city",
    schema: z.object({
      city: z.string().describe("The city to get the weather for"),
    }),
  }
);

const agent = createAgent({
  model: "gpt-5.5",
  tools: [getWeather],
});

console.log(
  await agent.invoke({
    messages: [{ role: "user", content: "What's the weather in San Francisco?" }],
  })
);

  8. Configure model providers for createAgent

    v0.1
    The createAgent function supports multiple model providers via specific string identifiers. Ensure the corresponding provider package is installed (e.g., @langchain/openai, @langchain/google-genai, @langchain/anthropic).
    // Model identifier examples for createAgent:

// OpenAI
model: "gpt-5.5"

// Google Gemini
model: "google-genai:gemini-2.5-flash-lite"

// Anthropic
model: "claude-sonnet-4-6"

// OpenRouter
model: "openrouter:anthropic/claude-sonnet-4-6"

// Fireworks
model: "fireworks:accounts/fireworks/models/qwen3p5-397b-a17b"

// Baseten
model: "baseten:zai-org/GLM-5.2"

// Ollama
model: "ollama:devstral-2"

// Azure OpenAI
model: "azure_openai:gpt-5.5"

// AWS Bedrock
model: "bedrock:gpt-5.5"