Overview

All functionality related to OpenAI

OpenAI is American artificial intelligence (AI) research laboratory consisting of the non-profit OpenAI Incorporated and its for-profit subsidiary corporation OpenAI Limited Partnership. OpenAI conducts AI research with the declared intention of promoting and developing a friendly AI. OpenAI systems run on an Azure-based supercomputing platform from Microsoft.

The OpenAI API is powered by a diverse set of models with different capabilities and price points.

ChatGPT is the Artificial Intelligence (AI) chatbot developed by OpenAI.

Installation and Setup

• Get an OpenAI api key and set it as an environment variable (OPENAI_API_KEY)

Chat model

See a usage example.

import { ChatOpenAI } from "@langchain/openai";

LLM

See a usage example.

See this section for general instructions on installing LangChain packages.

npm

npm install @langchain/openai @langchain/core

import { OpenAI } from "@langchain/openai";

Text Embedding Model

See a usage example

import { OpenAIEmbeddings } from "@langchain/openai";

Chain

import { OpenAIModerationChain } from "@langchain/classic/chains";

Middleware

Middleware specifically designed for OpenAI models. Learn more about middleware.

Middleware	Description
Content moderation	Moderate agent traffic using OpenAI’s moderation endpoint

Content moderation

Moderate agent traffic (user input, model output, and tool results) using OpenAI’s moderation endpoint to detect and handle unsafe content. Content moderation is useful for the following:

• Applications requiring content safety and compliance
• Filtering harmful, hateful, or inappropriate content
• Customer-facing agents that need safety guardrails
• Meeting platform moderation requirements

Learn more about OpenAI’s moderation models and categories.

API reference: openAIModerationMiddleware

import { createAgent, openAIModerationMiddleware } from "langchain";

const agent = createAgent({
  model: "openai:gpt-4o",
  tools: [searchTool, databaseTool],
  middleware: [
    openAIModerationMiddleware({
      model: "openai:gpt-4o",
      moderationModel: "omni-moderation-latest",
      checkInput: true,
      checkOutput: true,
      exitBehavior: "end",
    }),
  ],
});

Configuration options

model · string | BaseChatModel · required

OpenAI model to use for moderation. Can be either a model name string (e.g., "openai:gpt-4o") or a BaseChatModel instance. The middleware will use this model’s client to access the moderation endpoint.

moderationModel · ModerationModel · default: "omni-moderation-latest"

OpenAI moderation model to use. Options: 'omni-moderation-latest', 'omni-moderation-2024-09-26', 'text-moderation-latest', 'text-moderation-stable'

checkInput · boolean · default: "true"

Whether to check user input messages before the model is called

checkOutput · boolean · default: "true"

Whether to check model output messages after the model is called

checkToolResults · boolean · default: "false"

Whether to check tool result messages before the model is called

exitBehavior · 'error' | 'end' | 'replace' · default: "'end'"

How to handle violations when content is flagged. Options:

• 'end' - End agent execution immediately with a violation message
• 'error' - Throw OpenAIModerationError exception
• 'replace' - Replace the flagged content with the violation message and continue

violationMessage · string | undefined

Custom template for violation messages. Supports template variables:

• {categories} - Comma-separated list of flagged categories
• {category_scores} - JSON string of category scores
• {original_content} - The original flagged content

Default: "I'm sorry, but I can't comply with that request. It was flagged for {categories}."

Full example

The middleware integrates OpenAI’s moderation endpoint to check content at different stages:

Moderation stages:

• checkInput - User messages before model call
• checkOutput - AI messages after model call
• checkToolResults - Tool outputs before model call

Exit behaviors:

• 'end' (default) - Stop execution with violation message
• 'error' - Throw exception for application handling
• 'replace' - Replace flagged content and continue

import { createAgent, openAIModerationMiddleware } from "langchain";

// Basic moderation
const agent = createAgent({
  model: "openai:gpt-4o",
  tools: [searchTool, customerDataTool],
  middleware: [
    openAIModerationMiddleware({
      model: "openai:gpt-4o",
      moderationModel: "omni-moderation-latest",
      checkInput: true,
      checkOutput: true,
    }),
  ],
});

// Strict moderation with custom message
const agentStrict = createAgent({
  model: "openai:gpt-4o",
  tools: [searchTool, customerDataTool],
  middleware: [
    openAIModerationMiddleware({
      model: "openai:gpt-4o",
      moderationModel: "omni-moderation-latest",
      checkInput: true,
      checkOutput: true,
      checkToolResults: true,
      exitBehavior: "error",
      violationMessage:
        "Content policy violation detected: {categories}. " +
        "Please rephrase your request.",
    }),
  ],
});

// Moderation with replacement behavior
const agentReplace = createAgent({
  model: "openai:gpt-4o",
  tools: [searchTool],
  middleware: [
    openAIModerationMiddleware({
      model: "openai:gpt-4o",
      checkInput: true,
      exitBehavior: "replace",
      violationMessage: "[Content removed due to safety policies]",
    }),
  ],
});

---

Edit the source of this page on GitHub.

Connect these docs programmatically to Claude, VSCode, and more via MCP for real-time answers.