OpenAI Adapter | TanStack AI Docs

The OpenAI adapter provides access to OpenAI's GPT models, including GPT-4, GPT-3.5, and more.

Installation

bash

npm install @tanstack/ai-openai

npm install @tanstack/ai-openai

Basic Usage

typescript

import { chat } from "@tanstack/ai";
import { openai } from "@tanstack/ai-openai";

const adapter = openai({
  apiKey: process.env.OPENAI_API_KEY!,
});

const stream = chat({
  adapter,
  messages: [{ role: "user", content: "Hello!" }],
  model: "gpt-4o",
});

import { chat } from "@tanstack/ai";
import { openai } from "@tanstack/ai-openai";

const adapter = openai({
  apiKey: process.env.OPENAI_API_KEY!,
});

const stream = chat({
  adapter,
  messages: [{ role: "user", content: "Hello!" }],
  model: "gpt-4o",
});

Configuration

typescript

import { openai, type OpenAIConfig } from "@tanstack/ai-openai";

const config: OpenAIConfig = {
  apiKey: process.env.OPENAI_API_KEY!,
  organization: "org-...", // Optional
  baseURL: "https://api.openai.com/v1", // Optional, for custom endpoints
};

const adapter = openai(config);

import { openai, type OpenAIConfig } from "@tanstack/ai-openai";

const config: OpenAIConfig = {
  apiKey: process.env.OPENAI_API_KEY!,
  organization: "org-...", // Optional
  baseURL: "https://api.openai.com/v1", // Optional, for custom endpoints
};

const adapter = openai(config);

Available Models

Chat Models

gpt-4o - Latest GPT-4 model
gpt-4o-mini - Faster, cheaper GPT-4 variant
gpt-4-turbo - GPT-4 Turbo
gpt-4 - GPT-4 base model
gpt-3.5-turbo - GPT-3.5 Turbo
And many more...

Image Models

dall-e-3 - DALL-E 3 image generation
dall-e-2 - DALL-E 2 image generation

Embedding Models

text-embedding-3-large - Large embedding model
text-embedding-3-small - Small embedding model
text-embedding-ada-002 - Ada embedding model

Example: Chat Completion

typescript

import { chat, toStreamResponse } from "@tanstack/ai";
import { openai } from "@tanstack/ai-openai";

const adapter = openai({
  apiKey: process.env.OPENAI_API_KEY!,
});

export async function POST(request: Request) {
  const { messages } = await request.json();

  const stream = chat({
    adapter,
    messages,
    model: "gpt-4o",
  });

  return toStreamResponse(stream);
}

import { chat, toStreamResponse } from "@tanstack/ai";
import { openai } from "@tanstack/ai-openai";

const adapter = openai({
  apiKey: process.env.OPENAI_API_KEY!,
});

export async function POST(request: Request) {
  const { messages } = await request.json();

  const stream = chat({
    adapter,
    messages,
    model: "gpt-4o",
  });

  return toStreamResponse(stream);
}

Example: With Tools

typescript

import { chat, toolDefinition } from "@tanstack/ai";
import { openai } from "@tanstack/ai-openai";
import { z } from "zod";

const adapter = openai({
  apiKey: process.env.OPENAI_API_KEY!,
});

const getWeatherDef = toolDefinition({
  name: "get_weather",
  description: "Get the current weather",
  inputSchema: z.object({
    location: z.string(),
  }),
});

const getWeather = getWeatherDef.server(async ({ location }) => {
  // Fetch weather data
  return { temperature: 72, conditions: "sunny" };
});

const stream = chat({
  adapter,
  messages,
  model: "gpt-4o",
  tools: [getWeather],
});

import { chat, toolDefinition } from "@tanstack/ai";
import { openai } from "@tanstack/ai-openai";
import { z } from "zod";

const adapter = openai({
  apiKey: process.env.OPENAI_API_KEY!,
});

const getWeatherDef = toolDefinition({
  name: "get_weather",
  description: "Get the current weather",
  inputSchema: z.object({
    location: z.string(),
  }),
});

const getWeather = getWeatherDef.server(async ({ location }) => {
  // Fetch weather data
  return { temperature: 72, conditions: "sunny" };
});

const stream = chat({
  adapter,
  messages,
  model: "gpt-4o",
  tools: [getWeather],
});

Provider Options

OpenAI supports various provider-specific options:

typescript

const stream = chat({
  adapter: openai({ apiKey: process.env.OPENAI_API_KEY! }),
  messages,
  model: "gpt-4o",
  providerOptions: {
    temperature: 0.7,
    maxTokens: 1000,
    topP: 0.9,
    frequencyPenalty: 0.5,
    presencePenalty: 0.5,
  },
});

const stream = chat({
  adapter: openai({ apiKey: process.env.OPENAI_API_KEY! }),
  messages,
  model: "gpt-4o",
  providerOptions: {
    temperature: 0.7,
    maxTokens: 1000,
    topP: 0.9,
    frequencyPenalty: 0.5,
    presencePenalty: 0.5,
  },
});

Reasoning

Enable reasoning for models that support it (e.g., GPT-5). This allows the model to show its reasoning process, which is streamed as thinking chunks:

typescript

providerOptions: {
  reasoning: {
    effort: "medium", // "minimal" | "low" | "medium" | "high"
  },
}

providerOptions: {
  reasoning: {
    effort: "medium", // "minimal" | "low" | "medium" | "high"
  },
}

Supported Models:

gpt-5 - Supports reasoning with configurable effort
o3, o3-pro, o3-mini - Support reasoning

When reasoning is enabled, the model's reasoning process is streamed separately from the response text and appears as a collapsible thinking section in the UI.