fix(deps): update dependency @ai-sdk/openai to v3#118
Open
renovate[bot] wants to merge 1 commit intomainfrom
Open
fix(deps): update dependency @ai-sdk/openai to v3#118renovate[bot] wants to merge 1 commit intomainfrom
renovate[bot] wants to merge 1 commit intomainfrom
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
67fc8fb to
38e7eed
Compare
38e7eed to
0c9a548
Compare
0c9a548 to
0ce5f85
Compare
0ce5f85 to
3f96886
Compare
3f96886 to
050867e
Compare
050867e to
a5ba60e
Compare
a5ba60e to
7e67967
Compare
7e67967 to
9f274fd
Compare
9f274fd to
5307095
Compare
5307095 to
456bfc6
Compare
456bfc6 to
5c7099f
Compare
5c7099f to
85c245a
Compare
85c245a to
cfe51c1
Compare
cfe51c1 to
dd3d104
Compare
db60992 to
9d24462
Compare
9d24462 to
70405c1
Compare
70405c1 to
1cbc4f8
Compare
1cbc4f8 to
d9c6cdc
Compare
d9c6cdc to
4e081ad
Compare
4e081ad to
6ffa819
Compare
6ffa819 to
7d75459
Compare
7d75459 to
9c5aff5
Compare
9c5aff5 to
f74a247
Compare
f74a247 to
7b41859
Compare
7b41859 to
f3717cd
Compare
f3717cd to
840f995
Compare
840f995 to
179654b
Compare
179654b to
d6ea402
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR contains the following updates:
2.0.89→3.0.61Release Notes
vercel/ai (@ai-sdk/openai)
v3.0.61Compare Source
Patch Changes
b93f9b4: feat(provider/openai): forward imageDetail providerOptions on tool-result image contentv3.0.60Compare Source
Patch Changes
6dcd8e6: feat(openai): add GPT-5.5 chat model IDsv3.0.59Compare Source
Patch Changes
38966ab: fix(openai, openai-compatible): only send null content for assistant messages with tool callsv3.0.58Compare Source
Patch Changes
2370948: feat(openai): preservenamespaceon function_call output itemsv3.0.57Compare Source
Patch Changes
d33e7cc: chore(provider/openai): add type for image model options for type-safe processingv3.0.55Compare Source
Patch Changes
a727da4: chore: ensure consistent import handling and avoid import duplicates or cyclesa727da4]v3.0.54Compare Source
Patch Changes
a7f3c72: trigger release for all packages after provenance setup408a2ad: patch - send content: null instead of empty string for tool-only assistant messagesc71ad14: feat(provider/openai): add gpt-image-2 model supporta7f3c72]v3.0.53Compare Source
Patch Changes
953385d: fix(openai): default undefined tool-call input to empty object before serializing tool argumentsv3.0.52Compare Source
Patch Changes
d42076d: Add AI Gateway hint to provider READMEsv3.0.51Patch Changes
6247886]v3.0.50Patch Changes
0469aed]v3.0.49Patch Changes
bc01093: fix(openai): support file-url parts in tool output contentv3.0.48Patch Changes
9c548de: Addgpt-5.4-mini,gpt-5.4-mini-2026-03-17,gpt-5.4-nano, andgpt-5.4-nano-2026-03-17models.bcb04df: fix(openai): preserve raw finish reason for failed responses stream eventsHandle
response.failedchunks in Responses API streaming sofinishReason.rawis preserved fromincomplete_details.reason(e.g.max_output_tokens), and map failed-without-reason cases to unifiederrorinstead ofother.v3.0.47Patch Changes
055cd68: fix: publish v6 to latest npm dist tag055cd68]v3.0.46Compare Source
Patch Changes
75fc0e7: feat(openai): add new tool search toolv3.0.45Compare Source
Patch Changes
023088c: feat(provider/openai): addgpt-5.3-chat-latestv3.0.44Compare Source
Patch Changes
f4a734a: fix(provider/openai): drop reasoning parts without encrypted content when store: falsev3.0.43Compare Source
Patch Changes
64ac0fd]v3.0.42Compare Source
Patch Changes
2589004: feat(provider/openai): add GPT-5.4 model supportv3.0.41Compare Source
Patch Changes
ad4cfc2]v3.0.40Compare Source
Patch Changes
824b295]v3.0.39Compare Source
Patch Changes
08336f1]v3.0.38Compare Source
Patch Changes
64a8fae: chore: remove obsolete model IDs for Anthropic, Google, OpenAI, xAIv3.0.37Compare Source
Patch Changes
58bc42d: feat(provider/openai): support custom tools with alias mapping58bc42d]v3.0.36Compare Source
Patch Changes
53bdfa5: fix(openai): allow null/undefined type in streaming tool call deltasAzure AI Foundry and Mistral deployed on Azure omit the
typefield instreaming tool_calls deltas. The chat stream parser now accepts a missing
typefield (treating it as"function") instead of throwingInvalidResponseDataError: Expected 'function' type.Fixes #12770
v3.0.35Compare Source
Patch Changes
5e18272: fix(openai): include reasoning parts without itemId when encrypted_content is presentWhen
providerOptions.openai.itemIdis absent on a reasoning content part,the converter now uses
encrypted_contentas a fallback instead of silentlyskipping the part with a warning. The OpenAI Responses API accepts reasoning
items without an
idwhenencrypted_contentis supplied, enablingmulti-turn reasoning even when item IDs are stripped from provider options.
Also makes the
idfield optional on theOpenAIResponsesReasoningtype toreflect that the API does not require it.
Fixes #12853
v3.0.34Compare Source
Patch Changes
66a374c: Supportphaseparameter on Responses API message items. Thephasefield ('commentary'or'final_answer') is returned by models likegpt-5.3-codexon assistant message output items and must be preserved when sending follow-up requests. The phase value is available inproviderMetadata.openai.phaseon text parts and is automatically included on assistant messages sent back to the API.v3.0.33Compare Source
Patch Changes
624e651: Added missing model IDs to OpenAIChatModelId, OpenAIResponsesModelId, OpenAIImageModelId, OpenAISpeechModelId, OpenAITranscriptionModelId, and OpenAICompletionModelId types for better autocomplete support.v3.0.32Compare Source
Patch Changes
0c9395b: feat(provider/openai): addgpt-5.3-codexv3.0.31Compare Source
Patch Changes
d5f7312: fix(openai): change web search tool action to be optionalv3.0.30Compare Source
Patch Changes
ff12133: feat(provider/openai): support native skills and hosted shellv3.0.29Compare Source
Patch Changes
e2ee705: feat: differentiate text vs image input tokensv3.0.28Compare Source
Patch Changes
4024a3a]v3.0.27Compare Source
Patch Changes
99fbed8: feat: normalize provider specific model options type names and ensure they are exportedv3.0.26Compare Source
Patch Changes
7168375]v3.0.25Compare Source
Patch Changes
53f6731]v3.0.24Compare Source
Patch Changes
96936e5]v3.0.23Compare Source
Patch Changes
2810850]v3.0.22Compare Source
Patch Changes
1524271: chore: add skill information to README filesv3.0.21Compare Source
Patch Changes
2c70b90: chore: update provider docsv3.0.20Compare Source
Patch Changes
462ad00]v3.0.19Compare Source
Patch Changes
04c89b1: Provide Responses API providerMetadata types at the message / reasoning level.OpenaiResponsesProviderMetadataOpenaiResponsesReasoningProviderMetadataAzureResponsesProviderMetadataAzureResponsesReasoningProviderMetadatav3.0.18Compare Source
Patch Changes
4de5a1d: chore: excluded tests from src folder in npm package4de5a1d]v3.0.17Compare Source
Patch Changes
4218f86: fix(openai): preserve tool id for apply patch toolv3.0.16Compare Source
Patch Changes
2b8369d: chore: add docs to package distv3.0.15Compare Source
Patch Changes
8dc54db: chore: add src folders to package bundlev3.0.14Compare Source
Patch Changes
d21d016: feat(openai): add o4-mini model to OpenAIChatModelId typev3.0.13Compare Source
Patch Changes
000fa96: fix(openai): filter duplicate items when passing conversationIDv3.0.12Compare Source
Patch Changes
5c090e7]v3.0.11Compare Source
Patch Changes
46f46e4]v3.0.10Compare Source
Patch Changes
1b11dcb]v3.0.9Compare Source
Patch Changes
34d1c8a]v3.0.8Compare Source
Patch Changes
330bd92: Fix Responsescode_interpreterannotations and add typed providerMetadatacode_interpreterannotation types with the official spec.doGenerateanddoStreamstay in sync.OpenaiResponsesTextProviderMetadataOpenaiResponsesSourceDocumentProviderMetadataAzureResponsesTextProviderMetadataAzureResponsesSourceDocumentProviderMetadatav3.0.7Compare Source
Patch Changes
89202fb: fix(openai/azure): passing response_format correctlyv3.0.6Compare Source
Patch Changes
dc87517: Fix handling ofimage-urltool result content type in OpenAI Responses API conversionv3.0.5Compare Source
Patch Changes
d937c8f]v3.0.4Compare Source
Patch Changes
ad4cfc2]v3.0.3Compare Source
Patch Changes
08336f1]v3.0.2Compare Source
Patch Changes
e2ee705: feat: differentiate text vs image input tokensv3.0.1Compare Source
Patch Changes
04c89b1: Provide Responses API providerMetadata types at the message / reasoning level.OpenaiResponsesProviderMetadataOpenaiResponsesReasoningProviderMetadataAzureResponsesProviderMetadataAzureResponsesReasoningProviderMetadatav3.0.0Compare Source
Major Changes
dee8b05: ai SDK 6 betaMinor Changes
78928cb: release: start 5.1 betaPatch Changes
0c3b58b: fix(provider): add specificationVersion to ProviderV34920119: fix the "incomplete_details" key from nullable to nullish for openai compatibility0adc679: feat(provider): shared spec v392c6241: feat(openai): additional settings for file search tool88fc415: feat(openai): add the new provider 'apply_patch' tool817e601: fix(openai); fix url_citation schema in chat apidae2185: fix(openai): extract meta data from first chunk that contains any046aa3b: feat(provider): speech model v3 specf1277fe: feat(provider/openai): send assistant text and tool call parts as reference ids when store: true8d9e8ad: chore(provider): remove generics from EmbeddingModelV3Before
After
60f4775: fix: remove code for unsuported o1-mini and o1-preview models9a51b92: support OPENAI_BASE_URL envd64ece9: enables image_generation capabilities in the Azure provider through the Responses API.2625a04: feat(openai); update spec for mcp approval2e86082: feat(provider/openai):OpenAIChatLanguageModelOptionstype0877683: feat(provider/openai): support conversations apid0f1baf: feat(openai): Add support for 'promptCacheRetention: 24h' for gpt5.1 series831b6cc: feat(openai): adding provider mcp tool for openai95f65c2: chore: use import * from zod/v4edc5548: feat(provider/openai): automatically add reasoning.encrypted_content include when store = false954c356: feat(openai): allow custom names for provider-defined tools544d4e8: chore(specification): rename v3 provider defined tool to provider tool77f2b20: enables code_interpreter and file_search capabilities in the Azure provider through the Responses API0c4822d: feat:EmbeddingModelV373d9883: chore(openai): enable strict json by defaultd2039d7: feat(provider/openai): add GPT 5.1 Codex Max to OpenAI Responses model IDs list88edc28: feat (provider/openai): include more image generation response metadatae8109d3: feat: tool execution approvaled329cb: feat:Provider-V33bd2689: feat: extended token usage1cad0ab: feat: add provider version to user-agent headere85fa2f: feat(openai): add sources in web-search actions423ba08: Set the annotations from the Responses API to doStream401f561: fix(provider/openai): fix web search tool input types4122d2a: feat(provider/openai): add gpt-5-codex model id0153bfa: fix(openai): fix parameter exclusion logic8dac895: feat:LanguageModelV3304222e: Add streaming support for apply_patch partial diffs.23f132b: fix: error schema for Responses API1d0de66: refactoring(provider/openai): simplify code000e87b: fix(provider/openai): add providerExecuted flag to tool start chunks2c0a758: chore(openai): add JSDoc to responses options1b982e6: feat(openai): preserve file_id when converting file citationsb82987c: feat(openai): support openai code-interpreter annotations457318b: chore(provider,ai): switch to SharedV3Warning and unified warningsb681d7d: feat: expose usage tokens for 'generateImage' function79b4e46: feat(openai): add 'gpt-5.1' modelID3997a42: feat(provider/openai): local shell tool348fd10: fix(openai): treat unknown models as reasoning9061dc0: feat: image editingfe49278: feat(provider/openai): only send item references for reasoning when store: truecb4d238: The built in Code Interpreter tool input code is streamed intool-input-<start/delta/end>chunks.357cfd7: feat(provider/openai): add new model IDsgpt-image-1-mini,gpt-5-pro,gpt-5-pro-2025-10-0638a4035: added support for external_web_access parameter on web_search tool40d5419: feat(openai): addo3-deep-researchando4-mini-deep-researchmodels366f50b: chore(provider): add deprecated textEmbeddingModel and textEmbedding aliases2b0caef: feat(provider/openai): preview image generation resultsb60d2e2: fix(openai): allow open_page action type url to be nullishfd47df5: fix(openai): revised_prompt sometimes returns null, causing errors4616b86: chore: update zod peer depenedency version7756857: fix(provider/openai): add truncation parameter support for Responses APIcad6445: feat(openai); adding OpenAI's new shell tool64aa48f: Azure OpenAI enabled web-search-preview0b9fdd5: fix(provider/openai): end reasoning parts earlier61c52dc: feat (provider/openai): add gpt-image-1.5 model supportef739fa: fix(openai): refactor apply-patch tool3220329: 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.d270a5d: chore(openai): update tests for apply-patch tool to use snapshotsf18ef7f: feat(openai): add gpt-5.2 models21e20c0: feat(provider): transcription model v3 spec522f6b8: feat:ImageModelV3484aa93: Add 'default' as service tier88574c1: ChangeisReasoningModeldetection from blocklist to allowlist and add override option68c6187: feat(provider/openai): support file and image tool results3794514: feat: flexible tool output content supportcbf52cd: feat: expose raw finish reason10c1322: fix: moved dependency@ai-sdk/test-serverto devDependencies5648ec0: Add GPT-5.2 support for non-reasoning parameters (temperature, topP, logProbs) when reasoningEffort is none.78f813e: fix(openai): allow temperature etc setting when reasoning effort is none for gpt-5.140dc7fa: fix(openai): change find action type to find_in_page action type0273b74: fix(openai): add support for sources type 'api'5bf101a: feat(provider/openai): add support for OpenAI xhigh reasoning effort1bd7d32: feat: tool-specific strict moded86b52f: distinguish between OpenAI and Azure in Responses API providerMetadata95f65c2: chore: load zod schemas lazily59561f8: fix(openai); fix url_citation schema in chat apiUpdated dependencies
v2.0.106Compare Source
Patch Changes
3a7454a: chore(provider/openai): add type for image model options for type-safe processingv2.0.105Compare Source
Patch Changes
783fa6c: chore: ensure consistent import handling and avoid import duplicates or cycles783fa6c]v2.0.104Compare Source
Patch Changes
0a00b9b: trigger release for all packages after provenance setup11c7bdd: feat(provider/openai): add gpt-image-2 model support0a00b9b]v2.0.103Patch Changes
5543cd1: Add AI Gateway hint to provider READMEsv2.0.102Patch Changes
a27a978]v2.0.101Patch Changes
316517c: Addgpt-5.4-mini,gpt-5.4-mini-2026-03-17,gpt-5.4-nano, andgpt-5.4-nano-2026-03-17models.v2.0.100Compare Source
Patch Changes
216151a: fix(provider/openai): drop reasoning parts without encrypted content when store: falsev2.0.99Compare Source
Patch Changes
0aab803: feat(provider/openai): addgpt-5.3-chat-latestv2.0.98Compare Source
Patch Changes
392dc94: feat(provider/openai): add GPT-5.4 model supportv2.0.97Compare Source
Patch Changes
6a2f01b]17d64e3]v2.0.96Compare Source
Patch Changes
e867b5c: Supportphaseparameter on Responses API message items. Thephasefield ('commentary'or'final_answer') is returned by models likegpt-5.3-codexon assistant message output items and must be preserved when sending follow-up requests. The phase value is available inproviderMetadata.openai.phaseon text parts and is automatically included on assistant messages sent back to the API.v2.0.95Compare Source
Patch Changes
b08351e: fix(openai): allow null/undefined type in streaming tool call deltasAzure AI Foundry and Mistral deployed on Azure omit the
typefield instreaming tool_calls deltas. The chat stream parser now accepts a missing
typefield (treating it as"function") instead of throwingInvalidResponseDataError: Expected 'function' type.Fixes #12770
v2.0.94Compare Source
Patch Changes
42815ac: feat(provider/openai): addgpt-5.3-codexv2.0.93Compare Source
Patch Changes
fa64d70: fix(openai): change web search tool action to be optionalv2.0.92Compare Source
Patch Changes
c680a01: fix(openai): add changeset and tests for per-image usage metadatav2.0.91Compare Source
Patch Changes
6a0adb7: feat: differentiate text vs image input tokensv2.0.90Compare Source
Patch Changes
20565b8]Configuration
📅 Schedule: (in timezone Asia/Singapore)
* 0-3 1 * *)🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.
♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.
🔕 Ignore: Close this PR and you won't be reminded about this update again.
This PR was generated by Mend Renovate. View the repository job log.