@ai-sdk/openai

openai: Fix parameter exclusion logic

openai: Allow temperature etc setting when reasoning effort is none for gpt-5.1

openai: Allow open_page action type url to be nullish

Feat (provider/openai): include more image generation response metadata

Distinguish between OpenAI and Azure in Responses API providerMetadata

refactoring(provider/openai): simplify code

Added support for external_web_access parameter on web_search tool

Azure OpenAI enabled web-search-preview

Set the annotations from the Responses API to doStream

Error schema for Responses API

provider/openai: Support conversations api

openai: Add support for 'promptCacheRetention: 24h' for gpt5.1 series

openai: Add 'gpt-5.1' modelID

openai: Add o3-deep-research and o4-mini-deep-research models

openai: Extract meta data from first chunk that contains any

backport: Test server

openai: Add support for sources type 'api'

Remove code for unsuported o1-mini and o1-preview models

openai: revised_prompt sometimes returns null, causing errors

openai: Preserve file_id when converting file citations

openai: Add sources in web-search actions

provider/openai: Add truncation parameter support for Responses API

provider/openai: End reasoning parts earlier

provider/openai: Fix web search tool input types

provider/openai: Send assistant text and tool call parts as reference ids when store: true

provider/openai: Automatically add reasoning.encrypted_content include when store = false

provider/openai: Support file and image tool results

Add 'default' as service tier

provider/openai: Add providerExecuted flag to tool start chunks

provider/openai: Add new model IDs gpt-image-1-mini, gpt-5-pro, gpt-5-pro-2025-10-06

Support OPENAI_BASE_URL env

provider/openai: Add gpt-5-codex model id

provider/openai: Local shell tool

The built in Code Interpreter tool input code is streamed in tool-input-<start/delta/end> chunks.

Enables code_interpreter and file_search capabilities in the Azure provider through the Responses API

provider/openai: Only send item references for reasoning when store: true

provider/openai: OpenAIChatLanguageModelOptions type

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

await generateText({
  model: openai.chat('gpt-4o'),
  prompt: 'Invent a new holiday and describe its traditions.',
  providerOptions: {
    openai: {
      user: 'user-123',
    } satisfies OpenAIChatLanguageModelOptions,
  },
});

Add provider version to user-agent header

Fix the "incomplete_details" key from nullable to nullish for openai compatibility

provider/openai: Remove provider-executed tools from chat completions model

provider/openai: Rework file search tool

provider/openai: Add maxToolCalls provider option

provider/openai: Send item references for provider-executed tool results

Feat (provider/openai): provider defined image generation tool support