OpenAI Provider

The OpenAI provider contains language model support for the OpenAI responses, chat, and completion APIs, as well as embedding model support for the OpenAI embeddings API.

Setup

The OpenAI provider is available in the @ai-sdk/openai module. You can install it with

<Tabs items={['pnpm', 'npm', 'yarn', 'bun']}> <Tab> <Snippet text="pnpm add @ai-sdk/openai" dark /> </Tab> <Tab> <Snippet text="npm install @ai-sdk/openai" dark /> </Tab> <Tab> <Snippet text="yarn add @ai-sdk/openai" dark /> </Tab>

Provider Instance

You can import the default provider instance openai from @ai-sdk/openai:

import { openai } from '@ai-sdk/openai';

If you need a customized setup, you can import createOpenAI from @ai-sdk/openai and create a provider instance with your settings:

import { createOpenAI } from '@ai-sdk/openai';

const openai = createOpenAI({
  // custom settings, e.g.
  headers: {
    'header-name': 'header-value',
  },
});

You can use the following optional settings to customize the OpenAI provider instance:

• baseURL string

  Use a different URL prefix for API calls, e.g. to use proxy servers. The default prefix is https://api.openai.com/v1.

• apiKey string

  API key that is being sent using the Authorization header. It defaults to the OPENAI_API_KEY environment variable.

• name string

  The provider name. You can set this when using OpenAI compatible providers to change the model provider property. Defaults to openai.

• organization string

  OpenAI Organization.

• project string

  OpenAI project.

• headers Record<string,string>

  Custom headers to include in the requests.

• fetch (input: RequestInfo, init?: RequestInit) => Promise<Response>

  Custom fetch implementation. Defaults to the global fetch function. You can use it as a middleware to intercept requests, or to provide a custom fetch implementation for e.g. testing.

Language Models

The OpenAI provider instance is a function that you can invoke to create a language model:

const model = openai('gpt-5');

It automatically selects the correct API based on the model id. You can also pass additional settings in the second argument:

const model = openai('gpt-5', {
  // additional settings
});

The available options depend on the API that's automatically chosen for the model (see below). If you want to explicitly select a specific model API, you can use .responses, .chat, or .completion.

Example

You can use OpenAI language models to generate text with the generateText function:

import { openai } from '@ai-sdk/openai';
import { generateText } from 'ai';

const { text } = await generateText({
  model: openai('gpt-5'),
  prompt: 'Write a vegetarian lasagna recipe for 4 people.',
});

OpenAI language models can also be used in the streamText function and support structured data generation with Output (see AI SDK Core).

Responses Models

You can use the OpenAI responses API with the openai(modelId) or openai.responses(modelId) factory methods. It is the default API that is used by the OpenAI provider (since AI SDK 5).

const model = openai('gpt-5');

Further configuration can be done using OpenAI provider options. You can validate the provider options using the OpenAILanguageModelResponsesOptions type.

import { openai, OpenAILanguageModelResponsesOptions } from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai('gpt-5'), // or openai.responses('gpt-5')
  providerOptions: {
    openai: {
      parallelToolCalls: false,
      store: false,
      user: 'user_123',
      // ...
    } satisfies OpenAILanguageModelResponsesOptions,
  },
  // ...
});

The following provider options are available:

• parallelToolCalls boolean Whether to use parallel tool calls. Defaults to true.

• store boolean

  Whether to store the generation. Defaults to true.

• maxToolCalls integer The maximum number of total calls to built-in tools that can be processed in a response. This maximum number applies across all built-in tool calls, not per individual tool. Any further attempts to call a tool by the model will be ignored.

• metadata Record<string, string> Additional metadata to store with the generation.

• conversation string The ID of the OpenAI Conversation to continue. You must create a conversation first via the OpenAI API. Cannot be used in conjunction with previousResponseId. Defaults to undefined.

• previousResponseId string The ID of the previous response. You can use it to continue a conversation. Defaults to undefined.

• instructions string Instructions for the model. They can be used to change the system or developer message when continuing a conversation using the previousResponseId option. Defaults to undefined.

• logprobs boolean | number Return the log probabilities of the tokens. Including logprobs will increase the response size and can slow down response times. However, it can be useful to better understand how the model is behaving. Setting to true returns the log probabilities of the tokens that were generated. Setting to a number (1-20) returns the log probabilities of the top n tokens that were generated.

• user string A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. Defaults to undefined.

• reasoningEffort 'none' | 'minimal' | 'low' | 'medium' | 'high' | 'xhigh' Reasoning effort for reasoning models. Defaults to medium. If you use providerOptions to set the reasoningEffort option, this model setting will be ignored.

• reasoningSummary 'auto' | 'detailed' Controls whether the model returns its reasoning process. Set to 'auto' for a condensed summary, 'detailed' for more comprehensive reasoning. Defaults to undefined (no reasoning summaries). When enabled, reasoning summaries appear in the stream as events with type 'reasoning' and in non-streaming responses within the reasoning field.

• strictJsonSchema boolean Whether to use strict JSON schema validation. Defaults to true.

• serviceTier 'auto' | 'flex' | 'priority' | 'default' Service tier for the request. Set to 'flex' for 50% cheaper processing at the cost of increased latency (available for o3, o4-mini, and gpt-5 models). Set to 'priority' for faster processing with Enterprise access (available for gpt-4, gpt-5, gpt-5-mini, o3, o4-mini; gpt-5-nano is not supported).

  Defaults to 'auto'.

• textVerbosity 'low' | 'medium' | 'high' Controls the verbosity of the model's response. Lower values result in more concise responses, while higher values result in more verbose responses. Defaults to 'medium'.

• include Array<string> Specifies additional content to include in the response. Supported values: ['file_search_call.results'] for including file search results in responses. ['message.output_text.logprobs'] for logprobs. Defaults to undefined.

• truncation string The truncation strategy to use for the model response.

  • Auto: If the input to this Response exceeds the model's context window size, the model will truncate the response to fit the context window by dropping items from the beginning of the conversation.
  • disabled (default): If the input size will exceed the context window size for a model, the request will fail with a 400 error.

• promptCacheKey string A cache key for manual prompt caching control. Used by OpenAI to cache responses for similar requests to optimize your cache hit rates.

• promptCacheRetention 'in_memory' | '24h' The retention policy for the prompt cache. Set to '24h' to enable extended prompt caching, which keeps cached prefixes active for up to 24 hours. Defaults to 'in_memory' for standard prompt caching. Note: '24h' is currently only available for the 5.1 series of models.

• safetyIdentifier string A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies. The IDs should be a string that uniquely identifies each user.

• systemMessageMode 'system' | 'developer' | 'remove' Controls the role of the system message when making requests. By default (when omitted), for models that support reasoning the system message is automatically converted to a developer message. Setting systemMessageMode to system passes the system message as a system-level instruction; developer passes it as a developer message; remove omits the system message from the request.

• forceReasoning boolean Force treating this model as a reasoning model. This is useful for "stealth" reasoning models (e.g. via a custom baseURL) where the model ID is not recognized by the SDK's allowlist. When enabled, the SDK applies reasoning-model parameter compatibility rules and defaults systemMessageMode to developer unless overridden.

• contextManagement Array<object> Enable server-side context management (compaction). When configured, the server automatically compresses conversation context when token usage crosses a specified threshold. Each object in the array should have:

  • type: 'compaction'
  • compactThreshold: number — the token count at which compaction is triggered

The OpenAI responses provider also returns provider-specific metadata:

For Responses models, you can type this metadata using OpenaiResponsesProviderMetadata:

import { openai, type OpenaiResponsesProviderMetadata } from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai('gpt-5'),
});

const providerMetadata = result.providerMetadata as
  | OpenaiResponsesProviderMetadata
  | undefined;

const { responseId, logprobs, serviceTier } = providerMetadata?.openai ?? {};

// responseId can be used to continue a conversation (previousResponseId).
console.log(responseId);

The following OpenAI-specific metadata may be returned:

• responseId string | null | undefined The ID of the response. Can be used to continue a conversation.
• logprobs (optional) Log probabilities of output tokens (when enabled).
• serviceTier (optional) Service tier information returned by the API.

Reasoning Output

For reasoning models like gpt-5, you can enable reasoning summaries to see the model's thought process. Different models support different summarizers—for example, o4-mini supports detailed summaries. Set reasoningSummary: "auto" to automatically receive the richest level available.

import {
  openai,
  type OpenAILanguageModelResponsesOptions,
} from '@ai-sdk/openai';
import { streamText } from 'ai';

const result = streamText({
  model: openai('gpt-5'),
  prompt: 'Tell me about the Mission burrito debate in San Francisco.',
  providerOptions: {
    openai: {
      reasoningSummary: 'detailed', // 'auto' for condensed or 'detailed' for comprehensive
    } satisfies OpenAILanguageModelResponsesOptions,
  },
});

for await (const part of result.fullStream) {
  if (part.type === 'reasoning') {
    console.log(`Reasoning: ${part.textDelta}`);
  } else if (part.type === 'text-delta') {
    process.stdout.write(part.textDelta);
  }
}

For non-streaming calls with generateText, the reasoning summaries are available in the reasoning field of the response:

import {
  openai,
  type OpenAILanguageModelResponsesOptions,
} from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai('gpt-5'),
  prompt: 'Tell me about the Mission burrito debate in San Francisco.',
  providerOptions: {
    openai: {
      reasoningSummary: 'auto',
    } satisfies OpenAILanguageModelResponsesOptions,
  },
});
console.log('Reasoning:', result.reasoning);

Learn more about reasoning summaries in the OpenAI documentation.

WebSocket Transport

OpenAI's WebSocket API keeps a persistent connection open, which can significantly reduce Time-to-First-Byte (TTFB) in agentic workflows with many tool calls. After the initial connection, subsequent requests skip TCP/TLS/HTTP negotiation entirely.

The ai-sdk-openai-websocket-fetch package provides a drop-in fetch replacement that routes streaming requests through a persistent WebSocket connection.

<Tabs items={['pnpm', 'npm', 'yarn', 'bun']}> <Tab> <Snippet text="pnpm add ai-sdk-openai-websocket-fetch" dark /> </Tab> <Tab> <Snippet text="npm install ai-sdk-openai-websocket-fetch" dark /> </Tab> <Tab> <Snippet text="yarn add ai-sdk-openai-websocket-fetch" dark /> </Tab> <Tab> <Snippet text="bun add ai-sdk-openai-websocket-fetch" dark /> </Tab> </Tabs>

Pass the WebSocket fetch to createOpenAI via the fetch option:

import { createOpenAI } from '@ai-sdk/openai';
import { createWebSocketFetch } from 'ai-sdk-openai-websocket-fetch';
import { streamText } from 'ai';

// Create a WebSocket-backed fetch instance
const wsFetch = createWebSocketFetch();
const openai = createOpenAI({ fetch: wsFetch });

const result = streamText({
  model: openai('gpt-4.1-mini'),
  prompt: 'Hello!',
  tools: {
    // ...
  },
  onFinish: () => wsFetch.close(), // close the WebSocket when done
});

The first request will be slower because it must establish the WebSocket connection (DNS + TCP + TLS + WebSocket upgrade). After that, subsequent steps in a multi-step tool-calling loop reuse the open connection, resulting in lower TTFB per step.

You can see a live side-by-side comparison of HTTP vs WebSocket streaming performance in the demo app.

Verbosity Control

You can control the length and detail of model responses using the textVerbosity parameter:

import {
  openai,
  type OpenAILanguageModelResponsesOptions,
} from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai('gpt-5-mini'),
  prompt: 'Write a poem about a boy and his first pet dog.',
  providerOptions: {
    openai: {
      textVerbosity: 'low', // 'low' for concise, 'medium' (default), or 'high' for verbose
    } satisfies OpenAILanguageModelResponsesOptions,
  },
});

The textVerbosity parameter scales output length without changing the underlying prompt:

• 'low': Produces terse, minimal responses
• 'medium': Balanced detail (default)
• 'high': Verbose responses with comprehensive detail

Web Search Tool

The OpenAI responses API supports web search through the openai.tools.webSearch tool.

const result = await generateText({
  model: openai('gpt-5'),
  prompt: 'What happened in San Francisco last week?',
  tools: {
    web_search: openai.tools.webSearch({
      // optional configuration:
      externalWebAccess: true,
      searchContextSize: 'high',
      userLocation: {
        type: 'approximate',
        city: 'San Francisco',
        region: 'California',
      },
      filters: {
        allowedDomains: ['sfchronicle.com', 'sfgate.com'],
      },
    }),
  },
  // Force web search tool (optional):
  toolChoice: { type: 'tool', toolName: 'web_search' },
});

// URL sources directly from `results`
const sources = result.sources;

// Or access sources from tool results
for (const toolResult of result.toolResults) {
  if (toolResult.toolName === 'web_search') {
    console.log('Query:', toolResult.output.action.query);
    console.log('Sources:', toolResult.output.sources);
    // `sources` is an array of object: { type: 'url', url: string }
  }
}

The web search tool supports the following configuration options:

• externalWebAccess boolean - Whether to use external web access for fetching live content. Defaults to true.
• searchContextSize 'low' | 'medium' | 'high' - Controls the amount of context used for the search. Higher values provide more comprehensive results but may have higher latency and cost.
• userLocation - Optional location information to provide geographically relevant results. Includes type (always 'approximate'), country, city, region, and timezone.
• filters - Optional filter configuration to restrict search results.
  • allowedDomains string[] - Array of allowed domains for the search. Subdomains of the provided domains are automatically included.

For detailed information on configuration options see the OpenAI Web Search Tool documentation.

File Search Tool

The OpenAI responses API supports file search through the openai.tools.fileSearch tool.

You can force the use of the file search tool by setting the toolChoice parameter to { type: 'tool', toolName: 'file_search' }.

const result = await generateText({
  model: openai('gpt-5'),
  prompt: 'What does the document say about user authentication?',
  tools: {
    file_search: openai.tools.fileSearch({
      vectorStoreIds: ['vs_123'],
      // configuration below is optional:
      maxNumResults: 5,
      filters: {
        key: 'author',
        type: 'eq',
        value: 'Jane Smith',
      },
      ranking: {
        ranker: 'auto',
        scoreThreshold: 0.5,
      },
    }),
  },
  providerOptions: {
    openai: {
      // optional: include results
      include: ['file_search_call.results'],
    } satisfies OpenAILanguageModelResponsesOptions,
  },
});

The file search tool supports filtering with both comparison and compound filters:

Comparison filters - Filter by a single attribute:

• eq - Equal to
• ne - Not equal to
• gt - Greater than
• gte - Greater than or equal to
• lt - Less than
• lte - Less than or equal to
• in - Value is in array
• nin - Value is not in array

// Single comparison filter
filters: { key: 'year', type: 'gte', value: 2023 }

// Filter with array values
filters: { key: 'status', type: 'in', value: ['published', 'reviewed'] }

Compound filters - Combine multiple filters with and or or:

// Compound filter with AND
filters: {
  type: 'and',
  filters: [
    { key: 'author', type: 'eq', value: 'Jane Smith' },
    { key: 'year', type: 'gte', value: 2023 },
  ],
}

// Compound filter with OR
filters: {
  type: 'or',
  filters: [
    { key: 'department', type: 'eq', value: 'Engineering' },
    { key: 'department', type: 'eq', value: 'Research' },
  ],
}

Image Generation Tool

OpenAI's Responses API supports multi-modal image generation as a provider-defined tool. Availability is restricted to specific models (for example, gpt-5 variants).

You can use the image tool with either generateText or streamText:

import { openai } from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai('gpt-5'),
  prompt:
    'Generate an image of an echidna swimming across the Mozambique channel.',
  tools: {
    image_generation: openai.tools.imageGeneration({ outputFormat: 'webp' }),
  },
});

for (const toolResult of result.staticToolResults) {
  if (toolResult.toolName === 'image_generation') {
    const base64Image = toolResult.output.result;
  }
}

import { openai } from '@ai-sdk/openai';
import { streamText } from 'ai';

const result = streamText({
  model: openai('gpt-5'),
  prompt:
    'Generate an image of an echidna swimming across the Mozambique channel.',
  tools: {
    image_generation: openai.tools.imageGeneration({
      outputFormat: 'webp',
      quality: 'low',
    }),
  },
});

for await (const part of result.fullStream) {
  if (part.type == 'tool-result' && !part.dynamic) {
    const base64Image = part.output.result;
  }
}

For complete details on model availability, image quality controls, supported sizes, and tool-specific parameters, refer to the OpenAI documentation:

• Image generation overview and models: OpenAI Image Generation
• Image generation tool parameters (background, size, quality, format, etc.): Image Generation Tool Options

Code Interpreter Tool

The OpenAI responses API supports the code interpreter tool through the openai.tools.codeInterpreter tool. This allows models to write and execute Python code.

import { openai } from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai('gpt-5'),
  prompt: 'Write and run Python code to calculate the factorial of 10',
  tools: {
    code_interpreter: openai.tools.codeInterpreter({
      // optional configuration:
      container: {
        fileIds: ['file-123', 'file-456'], // optional file IDs to make available
      },
    }),
  },
});

The code interpreter tool can be configured with:

• container: Either a container ID string or an object with fileIds to specify uploaded files that should be available to the code interpreter

MCP Tool

The OpenAI responses API supports connecting to Model Context Protocol (MCP) servers through the openai.tools.mcp tool. This allows models to call tools exposed by remote MCP servers or service connectors.

import { openai } from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai('gpt-5'),
  prompt: 'Search the web for the latest news about AI developments',
  tools: {
    mcp: openai.tools.mcp({
      serverLabel: 'web-search',
      serverUrl: 'https://mcp.exa.ai/mcp',
      serverDescription: 'A web-search API for AI agents',
    }),
  },
});

The MCP tool can be configured with:

• serverLabel string (required)

  A label to identify the MCP server. This label is used in tool calls to distinguish between multiple MCP servers.

• serverUrl string (required if connectorId is not provided)

  The URL for the MCP server. Either serverUrl or connectorId must be provided.

• connectorId string (required if serverUrl is not provided)

  Identifier for a service connector. Either serverUrl or connectorId must be provided.

• serverDescription string (optional)

  Optional description of the MCP server that helps the model understand its purpose.

• allowedTools string[] | object (optional)

  Controls which tools from the MCP server are available. Can be:

  • An array of tool names: ['tool1', 'tool2']
  • An object with filters:
    tsCopy

    {
      readOnly: true, // Only allow read-only tools
      toolNames: ['tool1', 'tool2'] // Specific tool names
    }
    

• authorization string (optional)

  OAuth access token for authenticating with the MCP server or connector.

• headers Record<string, string> (optional)

  Optional HTTP headers to include in requests to the MCP server.

• requireApproval 'always' | 'never' | object (optional)

  Controls which MCP tool calls require user approval before execution. Can be:

  • 'always': All MCP tool calls require approval
  • 'never': No MCP tool calls require approval (default)
  • An object with filters:
    tsCopy

    {
      never: {
        toolNames: ['safe_tool', 'another_safe_tool']; // Skip approval for these tools
      }
    }
    

  When approval is required, the model will return a tool-approval-request content part that you can use to prompt the user for approval. See Human in the Loop for more details on implementing approval workflows.

Local Shell Tool

The OpenAI responses API support the local shell tool for Codex models through the openai.tools.localShell tool. Local shell is a tool that allows agents to run shell commands locally on a machine you or the user provides.

import { openai } from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai.responses('gpt-5-codex'),
  tools: {
    local_shell: openai.tools.localShell({
      execute: async ({ action }) => {
        // ... your implementation, e.g. sandbox access ...
        return { output: stdout };
      },
    }),
  },
  prompt: 'List the files in my home directory.',
  stopWhen: stepCountIs(2),
});

Shell Tool

The OpenAI Responses API supports the shell tool through the openai.tools.shell tool. The shell tool allows running bash commands and interacting with a command line. The model proposes shell commands; your integration executes them and returns the outputs.

The shell tool supports three environment modes that control where commands are executed:

Local Execution (default)

When no environment is specified (or type: 'local' is used), commands are executed locally via your execute callback:

import { openai } from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai('gpt-5.2'),
  tools: {
    shell: openai.tools.shell({
      execute: async ({ action }) => {
        // ... your implementation, e.g. sandbox access ...
        return { output: results };
      },
    }),
  },
  prompt: 'List the files in the current directory and show disk usage.',
});

Hosted Container (auto)

Set environment.type to 'containerAuto' to run commands in an OpenAI-hosted container. No execute callback is needed — OpenAI handles execution server-side:

const result = await generateText({
  model: openai('gpt-5.2'),
  tools: {
    shell: openai.tools.shell({
      environment: {
        type: 'containerAuto',
        // optional configuration:
        memoryLimit: '4g',
        fileIds: ['file-abc123'],
        networkPolicy: {
          type: 'allowlist',
          allowedDomains: ['example.com'],
        },
      },
    }),
  },
  prompt: 'Install numpy and compute the eigenvalues of a 3x3 matrix.',
});

The containerAuto environment supports:

• fileIds string[] - File IDs to make available in the container
• memoryLimit '1g' | '4g' | '16g' | '64g' - Memory limit for the container
• networkPolicy - Network access policy:
  • { type: 'disabled' } — no network access
  • { type: 'allowlist', allowedDomains: string[], domainSecrets?: Array<{ domain, name, value }> } — allow specific domains with optional secrets

Existing Container Reference

Set environment.type to 'containerReference' to use an existing container by ID:

const result = await generateText({
  model: openai('gpt-5.2'),
  tools: {
    shell: openai.tools.shell({
      environment: {
        type: 'containerReference',
        containerId: 'cntr_abc123',
      },
    }),
  },
  prompt: 'Check the status of running processes.',
});

Execute Callback

For local execution (default or type: 'local'), your execute function must return an output array with results for each command:

• stdout string - Standard output from the command
• stderr string - Standard error from the command
• outcome - Either { type: 'timeout' } or { type: 'exit', exitCode: number }

Skills

Skills are versioned bundles of files with a SKILL.md manifest that extend the shell tool's capabilities. They can be attached to both containerAuto and local environments.

Container skills support two formats — by reference (for skills uploaded to OpenAI) or inline (as a base64-encoded zip):

const result = await generateText({
  model: openai('gpt-5.2'),
  tools: {
    shell: openai.tools.shell({
      environment: {
        type: 'containerAuto',
        skills: [
          // By reference:
          { type: 'skillReference', skillId: 'skill_abc123' },
          // Or inline:
          {
            type: 'inline',
            name: 'my-skill',
            description: 'What this skill does',
            source: {
              type: 'base64',
              mediaType: 'application/zip',
              data: readFileSync('./my-skill.zip').toString('base64'),
            },
          },
        ],
      },
    }),
  },
  prompt: 'Use the skill to solve this problem.',
});

Local skills point to a directory on disk containing a SKILL.md file:

const result = await generateText({
  model: openai('gpt-5.2'),
  tools: {
    shell: openai.tools.shell({
      execute: async ({ action }) => {
        // ... your local execution implementation ...
        return { output: results };
      },
      environment: {
        type: 'local',
        skills: [
          {
            name: 'my-skill',
            description: 'What this skill does',
            path: resolve('path/to/skill-directory'),
          },
        ],
      },
    }),
  },
  prompt: 'Use the skill to solve this problem.',
  stopWhen: stepCountIs(5),
});

For more details on creating skills, see the OpenAI Skills documentation.

Apply Patch Tool

The OpenAI Responses API supports the apply patch tool for GPT-5.1 models through the openai.tools.applyPatch tool. The apply patch tool lets the model create, update, and delete files in your codebase using structured diffs. Instead of just suggesting edits, the model emits patch operations that your application applies and reports back on, enabling iterative, multi-step code editing workflows.

import { openai } from '@ai-sdk/openai';
import { generateText, stepCountIs } from 'ai';

const result = await generateText({
  model: openai('gpt-5.1'),
  tools: {
    apply_patch: openai.tools.applyPatch({
      execute: async ({ callId, operation }) => {
        // ... your implementation for applying the diffs.
      },
    }),
  },
  prompt: 'Create a python file that calculates the factorial of a number',
  stopWhen: stepCountIs(5),
});

Your execute function must return:

• status 'completed' | 'failed' - Whether the patch was applied successfully
• output string (optional) - Human-readable log text (e.g., results or error messages)

Tool Search

Tool search allows the model to dynamically search for and load tools into context as needed, rather than loading all tool definitions up front. This can reduce token usage, cost, and latency when you have many tools. Mark the tools you want to make searchable with deferLoading: true in their providerOptions.

There are two execution modes:

• Server-executed (hosted): OpenAI searches across the deferred tools declared in the request and returns the loaded subset in the same response. No extra round-trip is needed.
• Client-executed: The model emits a tool_search_call, your application performs the lookup, and you return the matching tools via the execute callback.

Server-Executed (Hosted) Tool Search

Use hosted tool search when the candidate tools are already known at request time. Add openai.tools.toolSearch() with no arguments and mark your tools with deferLoading: true:

import { openai } from '@ai-sdk/openai';
import { generateText, tool, stepCountIs } from 'ai';
import { z } from 'zod';

const result = await generateText({
  model: openai.responses('gpt-5.4'),
  prompt: 'What is the weather in San Francisco?',
  stopWhen: stepCountIs(10),
  tools: {
    toolSearch: openai.tools.toolSearch(),

    get_weather: tool({
      description: 'Get the current weather at a specific location',
      inputSchema: z.object({
        location: z.string(),
        unit: z.enum(['celsius', 'fahrenheit']),
      }),
      execute: async ({ location, unit }) => ({
        location,
        temperature: unit === 'celsius' ? 18 : 64,
      }),
      providerOptions: {
        openai: { deferLoading: true },
      },
    }),

    search_files: tool({
      description: 'Search through files in the workspace',
      inputSchema: z.object({ query: z.string() }),
      execute: async ({ query }) => ({
        results: [`Found 3 files matching "${query}"`],
      }),
      providerOptions: {
        openai: { deferLoading: true },
      },
    }),
  },
});

In hosted mode, the model internally searches the deferred tools, loads the relevant ones, and proceeds to call them — all within a single response. The tool_search_call and tool_search_output items appear in the response with execution: 'server' and call_id: null.

Client-Executed Tool Search

Use client-executed tool search when tool discovery depends on runtime state — for example, tools that vary per tenant, project, or external system. Pass execution: 'client' along with a description, parameters schema, and an execute callback:

import { openai } from '@ai-sdk/openai';
import { generateText, tool, stepCountIs } from 'ai';
import { z } from 'zod';

const result = await generateText({
  model: openai.responses('gpt-5.4'),
  prompt: 'What is the weather in San Francisco?',
  stopWhen: stepCountIs(10),
  tools: {
    toolSearch: openai.tools.toolSearch({
      execution: 'client',
      description: 'Search for available tools based on what the user needs.',
      parameters: {
        type: 'object',
        properties: {
          goal: {
            type: 'string',
            description: 'What the user is trying to accomplish',
          },
        },
        required: ['goal'],
        additionalProperties: false,
      },
      execute: async ({ arguments: args }) => {
        // Your custom tool discovery logic here.
        // Return the tools that match the search goal.
        return {
          tools: [
            {
              type: 'function',
              name: 'get_weather',
              description: 'Get the current weather at a specific location',
              deferLoading: true,
              parameters: {
                type: 'object',
                properties: {
                  location: { type: 'string' },
                },
                required: ['location'],
                additionalProperties: false,
              },
            },
          ],
        };
      },
    }),

    get_weather: tool({
      description: 'Get the current weather at a specific location',
      inputSchema: z.object({ location: z.string() }),
      execute: async ({ location }) => ({
        location,
        temperature: 64,
        condition: 'Partly cloudy',
      }),
      providerOptions: {
        openai: { deferLoading: true },
      },
    }),
  },
});

In client mode, the flow spans two steps:

1. Step 1: The model emits a tool_search_call with execution: 'client' and a non-null call_id. The SDK calls your execute callback with the search arguments. Your callback returns the discovered tools.
2. Step 2: The SDK sends the tool_search_output (with the matching call_id) back to the model. The model can now call the loaded tools as normal function calls.

For more details, see the OpenAI Tool Search documentation.

Custom Tool

The OpenAI Responses API supports custom tools through the openai.tools.customTool tool. Custom tools return a raw string instead of JSON, optionally constrained to a grammar (regex or Lark syntax). This makes them useful for generating structured text like SQL queries, code snippets, or any output that must match a specific pattern.

import { openai } from '@ai-sdk/openai';
import { generateText, stepCountIs } from 'ai';

const result = await generateText({
  model: openai.responses('gpt-5.2-codex'),
  tools: {
    write_sql: openai.tools.customTool({
      description: 'Write a SQL SELECT query to answer the user question.',
      format: {
        type: 'grammar',
        syntax: 'regex',
        definition: 'SELECT .+',
      },
      execute: async input => {
        // input is a raw string matching the grammar, e.g. "SELECT * FROM users WHERE age > 25"
        const rows = await db.query(input);
        return JSON.stringify(rows);
      },
    }),
  },
  toolChoice: 'required',
  prompt: 'Write a SQL query to get all users older than 25.',
  stopWhen: stepCountIs(3),
});

Custom tools also work with streamText:

import { openai } from '@ai-sdk/openai';
import { streamText } from 'ai';

const result = streamText({
  model: openai.responses('gpt-5.2-codex'),
  tools: {
    write_sql: openai.tools.customTool({
      description: 'Write a SQL SELECT query to answer the user question.',
      format: {
        type: 'grammar',
        syntax: 'regex',
        definition: 'SELECT .+',
      },
    }),
  },
  toolChoice: 'required',
  prompt: 'Write a SQL query to get all users older than 25.',
});

for await (const chunk of result.fullStream) {
  if (chunk.type === 'tool-call') {
    console.log(`Tool: ${chunk.toolName}`);
    console.log(`Input: ${chunk.input}`);
  }
}

The custom tool can be configured with:

• description string (optional) - A description of what the tool does, to help the model understand when to use it.
• format object (optional) - The output format constraint. Omit for unconstrained text output.
  • type 'grammar' | 'text' - The format type. Use 'grammar' for constrained output or 'text' for explicit unconstrained text.
  • syntax 'regex' | 'lark' - (grammar only) The grammar syntax. Use 'regex' for regular expression patterns or 'lark' for Lark parser grammar.
  • definition string - (grammar only) The grammar definition string (a regex pattern or Lark grammar).
• execute function (optional) - An async function that receives the raw string input and returns a string result. Enables multi-turn tool calling.

Image Inputs

The OpenAI Responses API supports Image inputs for appropriate models. You can pass Image files as part of the message content using the 'image' type:

const result = await generateText({
  model: openai('gpt-5'),
  messages: [
    {
      role: 'user',
      content: [
        {
          type: 'text',
          text: 'Please describe the image.',
        },
        {
          type: 'image',
          image: readFileSync('./data/image.png'),
        },
      ],
    },
  ],
});

The model will have access to the image and will respond to questions about it. The image should be passed using the image field.

You can also pass a file-id from the OpenAI Files API.

{
  type: 'image',
  image: 'file-8EFBcWHsQxZV7YGezBC1fq'
}

You can also pass the URL of an image.

{
  type: 'image',
  image: 'https://sample.edu/image.png',
}

PDF Inputs

The OpenAI Responses API supports reading PDF files. You can pass PDF files as part of the message content using the file type:

const result = await generateText({
  model: openai('gpt-5'),
  messages: [
    {
      role: 'user',
      content: [
        {
          type: 'text',
          text: 'What is an embedding model?',
        },
        {
          type: 'file',
          data: readFileSync('./data/ai.pdf'),
          mediaType: 'application/pdf',
          filename: 'ai.pdf', // optional
        },
      ],
    },
  ],
});

You can also pass a file-id from the OpenAI Files API.

{
  type: 'file',
  data: 'file-8EFBcWHsQxZV7YGezBC1fq',
  mediaType: 'application/pdf',
}

You can also pass the URL of a pdf.

{
  type: 'file',
  data: 'https://sample.edu/example.pdf',
  mediaType: 'application/pdf',
  filename: 'ai.pdf', // optional
}

The model will have access to the contents of the PDF file and respond to questions about it. The PDF file should be passed using the data field, and the mediaType should be set to 'application/pdf'.

Structured Outputs

The OpenAI Responses API supports structured outputs. You can use generateText or streamText with Output to enforce structured outputs.

const result = await generateText({
  model: openai('gpt-4.1'),
  output: Output.object({
    schema: z.object({
      recipe: z.object({
        name: z.string(),
        ingredients: z.array(
          z.object({
            name: z.string(),
            amount: z.string(),
          }),
        ),
        steps: z.array(z.string()),
      }),
    }),
  }),
  prompt: 'Generate a lasagna recipe.',
});

Typed providerMetadata in Text Parts

When using the OpenAI Responses API, the SDK attaches OpenAI-specific metadata to output parts via providerMetadata.

This metadata can be used on the client side for tasks such as rendering citations or downloading files generated by the Code Interpreter. To enable type-safe handling of this metadata, the AI SDK exports dedicated TypeScript types.

For text parts, when part.type === 'text', the providerMetadata is provided in the form of OpenaiResponsesTextProviderMetadata.

This metadata includes the following fields:

• itemId The ID of the output item in the Responses API.

• annotations (optional) An array of annotation objects generated by the model. If no annotations are present, this property itself may be omitted (undefined).

  Each element in annotations is a discriminated union with a required type field. Supported types include, for example:

  • url_citation
  • file_citation
  • container_file_citation
  • file_path

  These annotations directly correspond to the annotation objects defined by the Responses API and can be used for inline reference rendering or output analysis. For details, see the official OpenAI documentation: Responses API – output text annotations.

import {
  openai,
  type OpenaiResponsesTextProviderMetadata,
} from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai('gpt-4.1-mini'),
  prompt:
    'Create a program that generates five random numbers between 1 and 100 with two decimal places, and show me the execution results. Also save the result to a file.',
  tools: {
    code_interpreter: openai.tools.codeInterpreter(),
    web_search: openai.tools.webSearch(),
    file_search: openai.tools.fileSearch({ vectorStoreIds: ['vs_1234'] }), // requires a configured vector store
  },
});

for (const part of result.content) {
  if (part.type === 'text') {
    const providerMetadata = part.providerMetadata as
      | OpenaiResponsesTextProviderMetadata
      | undefined;
    if (!providerMetadata) continue;
    const { itemId: _itemId, annotations } = providerMetadata.openai;

    if (!annotations) continue;
    for (const annotation of annotations) {
      switch (annotation.type) {
        case 'url_citation':
          // url_citation is returned from web_search and provides:
          // properties: type, url, title, start_index and end_index
          break;
        case 'file_citation':
          // file_citation is returned from file_search and provides:
          // properties: type, file_id, filename and index
          break;
        case 'container_file_citation':
          // container_file_citation is returned from code_interpreter and provides:
          // properties: type, container_id, file_id, filename, start_index and end_index
          break;
        case 'file_path':
          // file_path provides:
          // properties: type, file_id and index
          break;
        default: {
          const _exhaustiveCheck: never = annotation;
          throw new Error(
            `Unhandled annotation: ${JSON.stringify(_exhaustiveCheck)}`,
          );
        }
      }
    }
  }
}

Typed providerMetadata in Reasoning Parts

When using the OpenAI Responses API, reasoning output parts can include provider metadata. To handle this metadata in a type-safe way, use OpenaiResponsesReasoningProviderMetadata.

For reasoning parts, when part.type === 'reasoning', the providerMetadata is provided in the form of OpenaiResponsesReasoningProviderMetadata.

This metadata includes the following fields:

• itemId
  The ID of the reasoning item in the Responses API.
• reasoningEncryptedContent (optional)
  Encrypted reasoning content (only returned when requested via include: ['reasoning.encrypted_content']).

import {
  openai,
  type OpenaiResponsesReasoningProviderMetadata,
  type OpenAILanguageModelResponsesOptions,
} from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai('gpt-5'),
  prompt: 'How many "r"s are in the word "strawberry"?',
  providerOptions: {
    openai: {
      store: false,
      include: ['reasoning.encrypted_content'],
    } satisfies OpenAILanguageModelResponsesOptions,
  },
});

for (const part of result.content) {
  if (part.type === 'reasoning') {
    const providerMetadata = part.providerMetadata as
      | OpenaiResponsesReasoningProviderMetadata
      | undefined;

    const { itemId, reasoningEncryptedContent } =
      providerMetadata?.openai ?? {};
    console.log(itemId, reasoningEncryptedContent);
  }
}

Typed providerMetadata in Source Document Parts

For source document parts, when part.type === 'source' and sourceType === 'document', the providerMetadata is provided as OpenaiResponsesSourceDocumentProviderMetadata.

This metadata is also a discriminated union with a required type field. Supported types include:

• file_citation
• container_file_citation
• file_path

Each type includes the identifiers required to work with the referenced resource, such as fileId and containerId.

import {
  openai,
  type OpenaiResponsesSourceDocumentProviderMetadata,
} from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai('gpt-4.1-mini'),
  prompt:
    'Create a program that generates five random numbers between 1 and 100 with two decimal places, and show me the execution results. Also save the result to a file.',
  tools: {
    code_interpreter: openai.tools.codeInterpreter(),
    web_search: openai.tools.webSearch(),
    file_search: openai.tools.fileSearch({ vectorStoreIds: ['vs_1234'] }), // requires a configured vector store
  },
});

for (const part of result.content) {
  if (part.type === 'source') {
    if (part.sourceType === 'document') {
      const providerMetadata = part.providerMetadata as
        | OpenaiResponsesSourceDocumentProviderMetadata
        | undefined;
      if (!providerMetadata) continue;
      const annotation = providerMetadata.openai;
      switch (annotation.type) {
        case 'file_citation':
          // file_citation is returned from file_search and provides:
          // properties: type, fileId and index
          // The filename can be accessed via part.filename.
          break;
        case 'container_file_citation':
          // container_file_citation is returned from code_interpreter and provides:
          // properties: type, containerId and fileId
          // The filename can be accessed via part.filename.
          break;
        case 'file_path':
          // file_path provides:
          // properties: type, fileId and index
          break;
        default: {
          const _exhaustiveCheck: never = annotation;
          throw new Error(
            `Unhandled annotation: ${JSON.stringify(_exhaustiveCheck)}`,
          );
        }
      }
    }
  }
}

Compaction

The OpenAI Responses API supports server-side context compaction. When enabled, the server automatically compresses conversation context when token usage crosses a configured threshold. This is useful for long-running conversations or agent loops where you want to stay within token limits without manually managing context.

The compaction item returned by the server is opaque and encrypted — it carries forward key prior state and reasoning into the next turn using fewer tokens. The AI SDK handles this automatically: compaction items are returned as text parts with special providerMetadata, and when passed back in subsequent requests they are sent as compaction input items.

import {
  openai,
  type OpenAILanguageModelResponsesOptions,
} from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai.responses('gpt-5.2'),
  messages: conversationHistory,
  providerOptions: {
    openai: {
      store: false,
      contextManagement: [{ type: 'compaction', compactThreshold: 50000 }],
    } satisfies OpenAILanguageModelResponsesOptions,
  },
});

Configuration:

• type — Must be 'compaction'
• compactThreshold — The token count at which compaction is triggered. When the rendered input token count crosses this threshold, the server runs a compaction pass before continuing inference.

Detecting Compaction in Streams

When using streamText, you can detect compaction by checking the providerMetadata on text-start and text-end events:

import {
  openai,
  type OpenAILanguageModelResponsesOptions,
} from '@ai-sdk/openai';
import { streamText } from 'ai';

const result = streamText({
  model: openai.responses('gpt-5.2'),
  messages: conversationHistory,
  providerOptions: {
    openai: {
      store: false,
      contextManagement: [{ type: 'compaction', compactThreshold: 50000 }],
    } satisfies OpenAILanguageModelResponsesOptions,
  },
});

for await (const part of result.fullStream) {
  switch (part.type) {
    case 'text-start': {
      const isCompaction = part.providerMetadata?.openai?.type === 'compaction';
      if (isCompaction) {
        // ... your logic
      }
      break;
    }
    case 'text-end': {
      const isCompaction = part.providerMetadata?.openai?.type === 'compaction';
      if (isCompaction) {
        // ... your logic
      }
      break;
    }
    case 'text-delta': {
      process.stdout.write(part.text);
      break;
    }
  }
}

Compaction in UI Applications

When using useChat or other UI hooks, compaction items appear as text parts with providerMetadata. You can detect and style them differently in your UI:

{
  message.parts.map((part, index) => {
    if (part.type === 'text') {
      const isCompaction =
        (part.providerMetadata?.openai as { type?: string } | undefined)
          ?.type === 'compaction';

      if (isCompaction) {
        return (
          <div
            key={index}
            className="bg-yellow-100 border-l-4 border-yellow-500 p-2"
          >
            <span className="font-bold">[Context Compacted]</span>
            <p className="text-sm text-yellow-700">
              The server compressed the conversation context to reduce token
              usage.
            </p>
          </div>
        );
      }
      return <div key={index}>{part.text}</div>;
    }
  });
}

The metadata includes the following fields:

• type — Always 'compaction'
• itemId string — The ID of the compaction item in the Responses API
• encryptedContent string (optional) — The encrypted compaction state. This is automatically sent back to the API when the message is included in subsequent requests.

Chat Models

You can create models that call the OpenAI chat API using the .chat() factory method. The first argument is the model id, e.g. gpt-4. The OpenAI chat models support tool calls and some have multi-modal capabilities.

const model = openai.chat('gpt-5');

OpenAI chat models support also some model specific provider options that are not part of the standard call settings. You can pass them in the providerOptions argument:

import { openai, type OpenAILanguageModelChatOptions } from '@ai-sdk/openai';

const model = openai.chat('gpt-5');

await generateText({
  model,
  providerOptions: {
    openai: {
      logitBias: {
        // optional likelihood for specific tokens
        '50256': -100,
      },
      user: 'test-user', // optional unique user identifier
    } satisfies OpenAILanguageModelChatOptions,
  },
});

The following optional provider options are available for OpenAI chat models:

• logitBias Record<number, number>

  Modifies the likelihood of specified tokens appearing in the completion.

  Accepts a JSON object that maps tokens (specified by their token ID in the GPT tokenizer) to an associated bias value from -100 to 100. You can use this tokenizer tool to convert text to token IDs. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.

  As an example, you can pass {"50256": -100} to prevent the token from being generated.

• logprobs boolean | number

  Return the log probabilities of the tokens. Including logprobs will increase the response size and can slow down response times. However, it can be useful to better understand how the model is behaving.

  Setting to true will return the log probabilities of the tokens that were generated.

  Setting to a number will return the log probabilities of the top n tokens that were generated.

• parallelToolCalls boolean

  Whether to enable parallel function calling during tool use. Defaults to true.

• user string

  A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. Learn more.

• reasoningEffort 'minimal' | 'low' | 'medium' | 'high' | 'xhigh'

  Reasoning effort for reasoning models. Defaults to medium. If you use providerOptions to set the reasoningEffort option, this model setting will be ignored.

• maxCompletionTokens number

  Maximum number of completion tokens to generate. Useful for reasoning models.

• store boolean

  Whether to enable persistence in Responses API.

• metadata Record<string, string>

  Metadata to associate with the request.

• prediction Record<string, any>

  Parameters for prediction mode.

• serviceTier 'auto' | 'flex' | 'priority' | 'default'

  Service tier for the request. Set to 'flex' for 50% cheaper processing at the cost of increased latency (available for o3, o4-mini, and gpt-5 models). Set to 'priority' for faster processing with Enterprise access (available for gpt-4, gpt-5, gpt-5-mini, o3, o4-mini; gpt-5-nano is not supported).

  Defaults to 'auto'.

• strictJsonSchema boolean

  Whether to use strict JSON schema validation. Defaults to true.

• textVerbosity 'low' | 'medium' | 'high'

  Controls the verbosity of the model's responses. Lower values will result in more concise responses, while higher values will result in more verbose responses.

• promptCacheKey string

  A cache key for manual prompt caching control. Used by OpenAI to cache responses for similar requests to optimize your cache hit rates.

• promptCacheRetention 'in_memory' | '24h'

  The retention policy for the prompt cache. Set to '24h' to enable extended prompt caching, which keeps cached prefixes active for up to 24 hours. Defaults to 'in_memory' for standard prompt caching. Note: '24h' is currently only available for the 5.1 series of models.

• safetyIdentifier string

  A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies. The IDs should be a string that uniquely identifies each user.

• systemMessageMode 'system' | 'developer' | 'remove'

  Override the system message mode for this model. If not specified, the mode is automatically determined based on the model. system uses the 'system' role for system messages (default for most models); developer uses the 'developer' role (used by reasoning models); remove removes system messages entirely.

• forceReasoning boolean

  Force treating this model as a reasoning model. This is useful for "stealth" reasoning models (e.g. via a custom baseURL) where the model ID is not recognized by the SDK's allowlist. When enabled, the SDK applies reasoning-model parameter compatibility rules and defaults systemMessageMode to developer unless overridden.

Reasoning

OpenAI has introduced the o1,o3, and o4 series of reasoning models. Currently, o4-mini, o3, o3-mini, and o1 are available via both the chat and responses APIs. The model gpt-5.1-codex-mini is available only via the responses API.

Reasoning models currently only generate text, have several limitations, and are only supported using generateText and streamText.

They support additional settings and response metadata:

• You can use providerOptions to set

  • the reasoningEffort option (or alternatively the reasoningEffort model setting), which determines the amount of reasoning the model performs.

• You can use response providerMetadata to access the number of reasoning tokens that the model generated.

import { openai, type OpenAILanguageModelChatOptions } from '@ai-sdk/openai';
import { generateText } from 'ai';

const { text, usage, providerMetadata } = await generateText({
  model: openai.chat('gpt-5'),
  prompt: 'Invent a new holiday and describe its traditions.',
  providerOptions: {
    openai: {
      reasoningEffort: 'low',
    } satisfies OpenAILanguageModelChatOptions,
  },
});

console.log(text);
console.log('Usage:', {
  ...usage,
  reasoningTokens: providerMetadata?.openai?.reasoningTokens,
});

• You can control how system messages are handled by providerOptions systemMessageMode:

  • developer: treat the prompt as a developer message (default for reasoning models).
  • system: keep the system message as a system-level instruction.
  • remove: remove the system message from the messages.

import { openai, type OpenAILanguageModelChatOptions } from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai.chat('gpt-5'),
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'Tell me a joke.' },
  ],
  providerOptions: {
    openai: {
      systemMessageMode: 'system',
    } satisfies OpenAILanguageModelChatOptions,
  },
});

Strict Structured Outputs

Strict structured outputs are enabled by default. You can disable them by setting the strictJsonSchema option to false.

import { openai, OpenAILanguageModelChatOptions } from '@ai-sdk/openai';
import { generateText, Output } from 'ai';
import { z } from 'zod';

const result = await generateText({
  model: openai.chat('gpt-4o-2024-08-06'),
  providerOptions: {
    openai: {
      strictJsonSchema: false,
    } satisfies OpenAILanguageModelChatOptions,
  },
  output: Output.object({
    schema: z.object({
      name: z.string(),
      ingredients: z.array(
        z.object({
          name: z.string(),
          amount: z.string(),
        }),
      ),
      steps: z.array(z.string()),
    }),
    schemaName: 'recipe',
    schemaDescription: 'A recipe for lasagna.',
  }),
  prompt: 'Generate a lasagna recipe.',
});

console.log(JSON.stringify(result.output, null, 2));

For example, optional schema properties are not supported. You need to change Zod .nullish() and .optional() to .nullable().

Logprobs

OpenAI provides logprobs information for completion/chat models. You can access it in the providerMetadata object.

import { openai, type OpenAILanguageModelChatOptions } from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai.chat('gpt-5'),
  prompt: 'Write a vegetarian lasagna recipe for 4 people.',
  providerOptions: {
    openai: {
      // this can also be a number,
      // refer to logprobs provider options section for more
      logprobs: true,
    } satisfies OpenAILanguageModelChatOptions,
  },
});

const openaiMetadata = (await result.providerMetadata)?.openai;

const logprobs = openaiMetadata?.logprobs;

Image Support

The OpenAI Chat API supports Image inputs for appropriate models. You can pass Image files as part of the message content using the 'image' type:

const result = await generateText({
  model: openai.chat('gpt-5'),
  messages: [
    {
      role: 'user',
      content: [
        {
          type: 'text',
          text: 'Please describe the image.',
        },
        {
          type: 'image',
          image: readFileSync('./data/image.png'),
        },
      ],
    },
  ],
});

The model will have access to the image and will respond to questions about it. The image should be passed using the image field.

You can also pass the URL of an image.

{
  type: 'image',
  image: 'https://sample.edu/image.png',
}

PDF support

The OpenAI Chat API supports reading PDF files. You can pass PDF files as part of the message content using the file type:

const result = await generateText({
  model: openai.chat('gpt-5'),
  messages: [
    {
      role: 'user',
      content: [
        {
          type: 'text',
          text: 'What is an embedding model?',
        },
        {
          type: 'file',
          data: readFileSync('./data/ai.pdf'),
          mediaType: 'application/pdf',
          filename: 'ai.pdf', // optional
        },
      ],
    },
  ],
});

The model will have access to the contents of the PDF file and respond to questions about it. The PDF file should be passed using the data field, and the mediaType should be set to 'application/pdf'.

You can also pass a file-id from the OpenAI Files API.

{
  type: 'file',
  data: 'file-8EFBcWHsQxZV7YGezBC1fq',
  mediaType: 'application/pdf',
}

You can also pass the URL of a PDF.

{
  type: 'file',
  data: 'https://sample.edu/example.pdf',
  mediaType: 'application/pdf',
  filename: 'ai.pdf', // optional
}

Predicted Outputs

OpenAI supports predicted outputs for gpt-4o and gpt-4o-mini. Predicted outputs help you reduce latency by allowing you to specify a base text that the model should modify. You can enable predicted outputs by adding the prediction option to the providerOptions.openai object:

const result = streamText({
  model: openai.chat('gpt-5'),
  messages: [
    {
      role: 'user',
      content: 'Replace the Username property with an Email property.',
    },
    {
      role: 'user',
      content: existingCode,
    },
  ],
  providerOptions: {
    openai: {
      prediction: {
        type: 'content',
        content: existingCode,
      },
    } satisfies OpenAILanguageModelChatOptions,
  },
});

OpenAI provides usage information for predicted outputs (acceptedPredictionTokens and rejectedPredictionTokens). You can access it in the providerMetadata object.

const openaiMetadata = (await result.providerMetadata)?.openai;

const acceptedPredictionTokens = openaiMetadata?.acceptedPredictionTokens;
const rejectedPredictionTokens = openaiMetadata?.rejectedPredictionTokens;

Image Detail

You can use the openai provider option to set the image input detail to high, low, or auto:

const result = await generateText({
  model: openai.chat('gpt-5'),
  messages: [
    {
      role: 'user',
      content: [
        { type: 'text', text: 'Describe the image in detail.' },
        {
          type: 'image',
          image:
            'https://github.com/vercel/ai/blob/main/examples/ai-functions/data/comic-cat.png?raw=true',

          // OpenAI specific options - image detail:
          providerOptions: {
            openai: { imageDetail: 'low' },
          },
        },
      ],
    },
  ],
});

Distillation

OpenAI supports model distillation for some models. If you want to store a generation for use in the distillation process, you can add the store option to the providerOptions.openai object. This will save the generation to the OpenAI platform for later use in distillation.

import { openai, type OpenAILanguageModelChatOptions } from '@ai-sdk/openai';
import { generateText } from 'ai';
import 'dotenv/config';

async function main() {
  const { text, usage } = await generateText({
    model: openai.chat('gpt-4o-mini'),
    prompt: 'Who worked on the original macintosh?',
    providerOptions: {
      openai: {
        store: true,
        metadata: {
          custom: 'value',
        },
      } satisfies OpenAILanguageModelChatOptions,
    },
  });

  console.log(text);
  console.log();
  console.log('Usage:', usage);
}

main().catch(console.error);

Prompt Caching

OpenAI has introduced Prompt Caching for supported models including gpt-4o and gpt-4o-mini.

• Prompt caching is automatically enabled for these models, when the prompt is 1024 tokens or longer. It does not need to be explicitly enabled.
• You can use response providerMetadata to access the number of prompt tokens that were a cache hit.
• Note that caching behavior is dependent on load on OpenAI's infrastructure. Prompt prefixes generally remain in the cache following 5-10 minutes of inactivity before they are evicted, but during off-peak periods they may persist for up to an hour.

import { openai } from '@ai-sdk/openai';
import { generateText } from 'ai';

const { text, usage, providerMetadata } = await generateText({
  model: openai.chat('gpt-4o-mini'),
  prompt: `A 1024-token or longer prompt...`,
});

console.log(`usage:`, {
  ...usage,
  cachedPromptTokens: providerMetadata?.openai?.cachedPromptTokens,
});

To improve cache hit rates, you can manually control caching using the promptCacheKey option:

import { openai, type OpenAILanguageModelChatOptions } from '@ai-sdk/openai';
import { generateText } from 'ai';

const { text, usage, providerMetadata } = await generateText({
  model: openai.chat('gpt-5'),
  prompt: `A 1024-token or longer prompt...`,
  providerOptions: {
    openai: {
      promptCacheKey: 'my-custom-cache-key-123',
    } satisfies OpenAILanguageModelChatOptions,
  },
});

console.log(`usage:`, {
  ...usage,
  cachedPromptTokens: providerMetadata?.openai?.cachedPromptTokens,
});

For GPT-5.1 models, you can enable extended prompt caching that keeps cached prefixes active for up to 24 hours:

import { openai, type OpenAILanguageModelChatOptions } from '@ai-sdk/openai';
import { generateText } from 'ai';

const { text, usage, providerMetadata } = await generateText({
  model: openai.chat('gpt-5.1'),
  prompt: `A 1024-token or longer prompt...`,
  providerOptions: {
    openai: {
      promptCacheKey: 'my-custom-cache-key-123',
      promptCacheRetention: '24h', // Extended caching for GPT-5.1
    } satisfies OpenAILanguageModelChatOptions,
  },
});

console.log(`usage:`, {
  ...usage,
  cachedPromptTokens: providerMetadata?.openai?.cachedPromptTokens,
});

Audio Input

With the gpt-4o-audio-preview model, you can pass audio files to the model.

import { openai } from '@ai-sdk/openai';
import { generateText } from 'ai';

const result = await generateText({
  model: openai.chat('gpt-4o-audio-preview'),
  messages: [
    {
      role: 'user',
      content: [
        { type: 'text', text: 'What is the audio saying?' },
        {
          type: 'file',
          mediaType: 'audio/mpeg',
          data: readFileSync('./data/galileo.mp3'),
        },
      ],
    },
  ],
});

Completion Models

You can create models that call the OpenAI completions API using the .completion() factory method. The first argument is the model id. Currently only gpt-3.5-turbo-instruct is supported.

const model = openai.completion('gpt-3.5-turbo-instruct');

OpenAI completion models support also some model specific settings that are not part of the standard call settings. You can pass them as an options argument:

const model = openai.completion('gpt-3.5-turbo-instruct');

await model.doGenerate({
  providerOptions: {
    openai: {
      echo: true, // optional, echo the prompt in addition to the completion
      logitBias: {
        // optional likelihood for specific tokens
        '50256': -100,
      },
      suffix: 'some text', // optional suffix that comes after a completion of inserted text
      user: 'test-user', // optional unique user identifier
    } satisfies OpenAILanguageModelCompletionOptions,
  },
});

The following optional provider options are available for OpenAI completion models:

• echo: boolean

  Echo back the prompt in addition to the completion.

• logitBias Record<number, number>

  Modifies the likelihood of specified tokens appearing in the completion.

  Accepts a JSON object that maps tokens (specified by their token ID in the GPT tokenizer) to an associated bias value from -100 to 100. You can use this tokenizer tool to convert text to token IDs. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.

  As an example, you can pass {"50256": -100} to prevent the <|endoftext|> token from being generated.

• logprobs boolean | number

  Return the log probabilities of the tokens. Including logprobs will increase the response size and can slow down response times. However, it can be useful to better understand how the model is behaving.

  Setting to true will return the log probabilities of the tokens that were generated.

  Setting to a number will return the log probabilities of the top n tokens that were generated.

• suffix string

  The suffix that comes after a completion of inserted text.

• user string

  A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. Learn more.

Model Capabilities

Embedding Models

You can create models that call the OpenAI embeddings API using the .embedding() factory method.

const model = openai.embedding('text-embedding-3-large');

OpenAI embedding models support several additional provider options. You can pass them as an options argument:

import { openai, type OpenAIEmbeddingModelOptions } from '@ai-sdk/openai';
import { embed } from 'ai';

const { embedding } = await embed({
  model: openai.embedding('text-embedding-3-large'),
  value: 'sunny day at the beach',
  providerOptions: {
    openai: {
      dimensions: 512, // optional, number of dimensions for the embedding
      user: 'test-user', // optional unique user identifier
    } satisfies OpenAIEmbeddingModelOptions,
  },
});

The following optional provider options are available for OpenAI embedding models:

• dimensions: number

  The number of dimensions the resulting output embeddings should have. Only supported in text-embedding-3 and later models.

• user string

  A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. Learn more.

Model Capabilities

Image Models

You can create models that call the OpenAI image generation API using the .image() factory method.

const model = openai.image('dall-e-3');

Image Editing

OpenAI's gpt-image-1 model supports powerful image editing capabilities. Pass input images via prompt.images to transform, combine, or edit existing images.

Basic Image Editing

Transform an existing image using text prompts:

const imageBuffer = readFileSync('./input-image.png');

const { images } = await generateImage({
  model: openai.image('gpt-image-1'),
  prompt: {
    text: 'Turn the cat into a dog but retain the style of the original image',
    images: [imageBuffer],
  },
});

Inpainting with Mask

Edit specific parts of an image using a mask. Transparent areas in the mask indicate where the image should be edited:

const image = readFileSync('./input-image.png');
const mask = readFileSync('./mask.png'); // Transparent areas = edit regions

const { images } = await generateImage({
  model: openai.image('gpt-image-1'),
  prompt: {
    text: 'A sunlit indoor lounge area with a pool containing a flamingo',
    images: [image],
    mask: mask,
  },
});

Background Removal

Remove the background from an image by setting background to transparent:

const imageBuffer = readFileSync('./input-image.png');

const { images } = await generateImage({
  model: openai.image('gpt-image-1'),
  prompt: {
    text: 'do not change anything',
    images: [imageBuffer],
  },
  providerOptions: {
    openai: {
      background: 'transparent',
      output_format: 'png',
    },
  },
});

Multi-Image Combining

Combine multiple reference images into a single output. gpt-image-1 supports up to 16 input images:

const cat = readFileSync('./cat.png');
const dog = readFileSync('./dog.png');
const owl = readFileSync('./owl.png');
const bear = readFileSync('./bear.png');

const { images } = await generateImage({
  model: openai.image('gpt-image-1'),
  prompt: {
    text: 'Combine these animals into a group photo, retaining the original style',
    images: [cat, dog, owl, bear],
  },
});

Model Capabilities

You can pass optional providerOptions to the image model. These are prone to change by OpenAI and are model dependent. For example, the gpt-image-1 model supports the quality option:

const { image, providerMetadata } = await generateImage({
  model: openai.image('gpt-image-1.5'),
  prompt: 'A salamander at sunrise in a forest pond in the Seychelles.',
  providerOptions: {
    openai: { quality: 'high' },
  },
});

For more on generateImage() see Image Generation.

OpenAI's image models return additional metadata in the response that can be accessed via providerMetadata.openai. The following OpenAI-specific metadata is available:

• images Array<object>

  Array of image-specific metadata. Each image object may contain:

  • revisedPrompt string - The revised prompt that was actually used to generate the image (OpenAI may modify your prompt for safety or clarity)
  • created number - The Unix timestamp (in seconds) of when the image was created
  • size string - The size of the generated image. One of 1024x1024, 1024x1536, or 1536x1024
  • quality string - The quality of the generated image. One of low, medium, or high
  • background string - The background parameter used for the image generation. Either transparent or opaque
  • outputFormat string - The output format of the generated image. One of png, webp, or jpeg

For more information on the available OpenAI image model options, see the OpenAI API reference.

Transcription Models

You can create models that call the OpenAI transcription API using the .transcription() factory method.

The first argument is the model id e.g. whisper-1.

const model = openai.transcription('whisper-1');

You can also pass additional provider-specific options using the providerOptions argument. For example, supplying the input language in ISO-639-1 (e.g. en) format will improve accuracy and latency.

import { experimental_transcribe as transcribe } from 'ai';
import { openai, type OpenAITranscriptionModelOptions } from '@ai-sdk/openai';

const result = await transcribe({
  model: openai.transcription('whisper-1'),
  audio: new Uint8Array([1, 2, 3, 4]),
  providerOptions: {
    openai: { language: 'en' } satisfies OpenAITranscriptionModelOptions,
  },
});

To get word-level timestamps, specify the granularity:

import { experimental_transcribe as transcribe } from 'ai';
import { openai, type OpenAITranscriptionModelOptions } from '@ai-sdk/openai';

const result = await transcribe({
  model: openai.transcription('whisper-1'),
  audio: new Uint8Array([1, 2, 3, 4]),
  providerOptions: {
    openai: {
      //timestampGranularities: ['word'],
      timestampGranularities: ['segment'],
    } satisfies OpenAITranscriptionModelOptions,
  },
});

// Access word-level timestamps
console.log(result.segments); // Array of segments with startSecond/endSecond

The following provider options are available:

• timestampGranularities string[] The granularity of the timestamps in the transcription. Defaults to ['segment']. Possible values are ['word'], ['segment'], and ['word', 'segment']. Note: There is no additional latency for segment timestamps, but generating word timestamps incurs additional latency.

• language string The language of the input audio. Supplying the input language in ISO-639-1 format (e.g. 'en') will improve accuracy and latency. Optional.

• prompt string An optional text to guide the model's style or continue a previous audio segment. The prompt should match the audio language. Optional.

• temperature number The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit. Defaults to 0. Optional.

• include string[] Additional information to include in the transcription response.

Model Capabilities

Speech Models

You can create models that call the OpenAI speech API using the .speech() factory method.

The first argument is the model id e.g. tts-1.

const model = openai.speech('tts-1');

The voice argument can be set to one of OpenAI's available voices: alloy, ash, coral, echo, fable, onyx, nova, sage, or shimmer.

import { experimental_generateSpeech as generateSpeech } from 'ai';
import { openai } from '@ai-sdk/openai';

const result = await generateSpeech({
  model: openai.speech('tts-1'),
  text: 'Hello, world!',
  voice: 'alloy', // OpenAI voice ID
});

You can also pass additional provider-specific options using the providerOptions argument:

import { experimental_generateSpeech as generateSpeech } from 'ai';
import { openai, type OpenAISpeechModelOptions } from '@ai-sdk/openai';

const result = await generateSpeech({
  model: openai.speech('tts-1'),
  text: 'Hello, world!',
  voice: 'alloy',
  providerOptions: {
    openai: {
      speed: 1.2,
    } satisfies OpenAISpeechModelOptions,
  },
});

• instructions string Control the voice of your generated audio with additional instructions e.g. "Speak in a slow and steady tone". Does not work with tts-1 or tts-1-hd. Optional.

• speed number The speed of the generated audio. Select a value from 0.25 to 4.0. Defaults to 1.0. Optional.

Model Capabilities