@ai-sdk/openai

Major Changes

• breaking

  Ai SDK 6 beta

Patch Changes

• provider: Add specificationVersion to ProviderV3

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

• provider: Shared spec v3

• openai: Additional settings for file search tool

• openai: Add the new provider 'apply_patch' tool

• fix(openai); fix url_citation schema in chat api

• openai: Extract meta data from first chunk that contains any

• provider: Speech model v3 spec

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

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

• Support OPENAI_BASE_URL env

• Enables image_generation capabilities in the Azure provider through the Responses API.

• feat(openai); update spec for mcp approval

• 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,
    },
  });

• provider/openai: Support conversations api

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

• openai: Adding provider mcp tool for openai

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

• openai: Allow custom names for provider-defined tools

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

• EmbeddingModelV3

• provider/openai: Add GPT 5.1 Codex Max to OpenAI Responses model IDs list

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

• Tool execution approval

• Provider-V3

• Extended token usage

• Add provider version to user-agent header

• openai: Add sources in web-search actions

• Set the annotations from the Responses API to doStream

• provider/openai: Fix web search tool input types

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

• openai: Fix parameter exclusion logic

• LanguageModelV3

• Add streaming support for apply_patch partial diffs.

• Error schema for Responses API

• refactoring(provider/openai): simplify code

• provider/openai: Add providerExecuted flag to tool start chunks

• openai: Preserve file_id when converting file citations

• openai: Support openai code-interpreter annotations

• Expose usage tokens for 'generateImage' function

• openai: Add 'gpt-5.1' modelID

• provider/openai: Local shell tool

• openai: Treat unknown models as reasoning

• Image editing

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

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

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

• Added support for external_web_access parameter on web_search tool

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

• provider/openai: Preview image generation results

• openai: Allow open_page action type url to be nullish

• openai: revised_prompt sometimes returns null, causing errors

• provider/openai: Add truncation parameter support for Responses API

• feat(openai); adding OpenAI's new shell tool

• Azure OpenAI enabled web-search-preview

• provider/openai: End reasoning parts earlier

• Feat (provider/openai): add gpt-image-1.5 model support

• openai: Refactor apply-patch tool

• Fix openai responses input: process all provider tool outputs (shell/apply_patch) so parallel tool results aren’t dropped and apply_patch outputs are forwarded.

• openai: Add gpt-5.2 models

• provider: Transcription model v3 spec

• ImageModelV3

• Add 'default' as service tier

• Change isReasoningModel detection from blocklist to allowlist and add override option

• provider/openai: Support file and image tool results

• Flexible tool output content support

• Expose raw finish reason

• Moved dependency @ai-sdk/test-server to devDependencies

• Add GPT-5.2 support for non-reasoning parameters (temperature, topP, logProbs) when reasoningEffort is none.

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

• openai: Change find action type to find_in_page action type

• openai: Add support for sources type 'api'

• provider/openai: Add support for OpenAI xhigh reasoning effort

• tool-specific strict mode

• Distinguish between OpenAI and Azure in Responses API providerMetadata

• fix(openai); fix url_citation schema in chat api