Full Changelog: v3.28.0...v3.29.0
- api: 5.4 nano and mini model slugs (c4ebb48)
- api: add in and nin values to comparison filter type (d060d44)
- internal: tweak CI branches (6379626)
Full Changelog: v3.27.0...v3.28.0
- api: add /v1/videos endpoint option to batch (7b2d67e)
- api: add defer_loading field to function tools (6d4b683)
- api: custom voices (d00b782)
- api: The
voiceparam and resouce has changed from astringto astring | {id: string}. This is a breaking change for Go.
Full Changelog: v3.26.0...v3.27.0
- api: add video character/edit/extend, remove Azure/webhook/polling/accumulator (20da043)
- api: add video edits/extensions/character, remove Azure/webhook/polling helpers (fa9413f)
- api: api update (c88c6c9)
- api: sora api improvements: character api, video extensions/edits, higher resolution exports. (93f6779)
- internal: codegen related update (d531232)
- internal: codegen related update (1748c11)
- internal: codegen related update (531ece1)
- internal: codegen related update (ce9f435)
Full Changelog: v3.25.0...v3.26.0
- api: The GA ComputerTool now uses the CompuerTool class. The 'computer_use_preview' tool is moved to ComputerUsePreview (347418b)
Full Changelog: v3.24.0...v3.25.0
- api: gpt-5.4, tool search tool, and new computer tool (101826d)
- api: remove Phase from input/output messages, PromptCacheKey from responses (961b8ca)
- internal: codegen related update (ab733b9)
- internal: codegen related update (23d1831)
- internal: reduce warnings (2963312)
Full Changelog: v3.23.0...v3.24.0
- api: add phase (72366d8)
- internal: move custom custom
jsontags toapi(0735303) - internal: refactor sse event parsing (45dc6bb)
Full Changelog: v3.23.0...v3.24.0
- api: add phase (72366d8)
- internal: move custom custom
jsontags toapi(0735303) - internal: refactor sse event parsing (45dc6bb)
Full Changelog: v3.22.1...v3.23.0
- api: add gpt-realtime-1.5 and gpt-audio-1.5 models to realtime session (9076e2f)
Full Changelog: v3.22.0...v3.22.1
- allow canceling a request while it is waiting to retry (54672cf)
- update mock server docs (3ac84dd)
- api: add batch size limit to vector store file batch parameters (f751c40)
- api: clarify safety_identifier max length in chat completions and responses (8257f9b)
- api: enhance method docstrings across audio/chat/completion/skill/upload/video APIs (38b8f63)
Full Changelog: v3.21.0...v3.22.0
- api: container network_policy and skills (8e5ea23)
- encoder: correctly serialize NullStruct (a6cb49e)
- update comment (bc316d7)
Full Changelog: v3.20.0...v3.21.0
- api: support for images in batch api (e23aeb1)
Full Changelog: v3.19.0...v3.20.0
- api: skills and hosted shell (9e191de)
Full Changelog: v3.18.0...v3.19.0
- api: responses context_management (199f230)
Full Changelog: v3.17.0...v3.18.0
- api: add shell_call_output status field (67a75d7)
- api: image generation actions for responses; ResponseFunctionCallArgumentsDoneEvent.name (2c57016)
- client: undo change to web search Find action (e340256)
- client: update type for
find_in_pageaction (4b5d499)
- client: improve example values (c86a65c)
- split
api.mdby standalone resources (aeed37b)
Full Changelog: v3.16.0...v3.17.0
- api: api update (a456c60)
- api: api updates (21fd4a9)
- client: add a convenient param.SetJSON helper (1b35ece)
- client: add a StreamError type to access raw events from sse streams (fd14b30)
- api: mark assistants as deprecated (9a8c9af)
- client: retain streaming when user sets request body (3a7a22e)
- docs: add missing pointer prefix to api.md return types (dd641d9)
Full Changelog: v3.15.0...v3.16.0
- api: add new Response completed_at prop (bff6331)
- client: use the correct order of params for vector store file and batch polling (ef32641)
- internal: codegen related update (1e529a4)
- internal: use different example values for some enums (a2836ee)
- update URL version (dc00e14)
Full Changelog: v3.14.0...v3.15.0
- rebuild (8205ae7)
- add float64 to valid types for RegisterFieldValidator (e67d89d)
Full Changelog: v3.13.0...v3.14.0
- api: gpt-image-1.5 (6102f02)
Full Changelog: v3.12.0...v3.13.0
- azure: correct Azure OpenAI API URL construction and auth (3ba3736)
Full Changelog: v3.11.0...v3.12.0
Full Changelog: v3.10.0...v3.11.0
- api: make model required for the responses/compact endpoint (05f8f4d)
Full Changelog: v3.9.0...v3.10.0
- api: gpt-5.1-codex-max and responses/compact (1e1ca2a)
Full Changelog: v3.8.1...v3.9.0
- api: gpt 5.1 (470f91f)
- api: align types of input items / output items for typescript (5b89d3b)
- client: correctly specify Accept header with / instead of empty (fbadb4e)
Full Changelog: v3.8.0...v3.8.1
- api: fix nullability of logprobs (b5aeb99)
Full Changelog: v3.7.0...v3.8.0
- api: Realtime API token_limits, Hybrid searching ranking options (9495f4a)
- internal: grammar fix (it's -> its) (879772d)
Full Changelog: v3.6.1...v3.7.0
- api: remove InputAudio from ResponseInputContent (cf50c53)
- azure: allow passing custom scopes (#541) (dffa08e)
- api: docs updates (94d54c1)
Full Changelog: v3.6.0...v3.6.1
- api: fix discriminator propertyName for ResponseFormatJsonSchema (57b0505)
Full Changelog: v3.5.0...v3.6.0
- api: Add responses.input_tokens.count (a43f2ce)
- api: internal openapi updates (7ad9b02)
Full Changelog: v3.4.0...v3.5.0
- api: api update (1aa78dd)
Full Changelog: v3.3.0...v3.4.0
- api: Add support for gpt-4o-transcribe-diarize on audio/transcriptions endpoint (ee32400)
- api: internal updates (74c8031)
- client: undo more naming changes (db441bc)
- client: undo some naming changes (a5aa3d6)
Full Changelog: v3.2.0...v3.3.0
- api: comparison filter in/not in (d6daca0)
Full Changelog: v3.1.0...v3.2.0
- api: dev day 2025 launches (d40a768)
Full Changelog: v3.0.1...v3.1.0
- api: add support for realtime calls (565ca67)
Full Changelog: v3.0.0...v3.0.1
- api: add status, approval_request_id to MCP tool call (a7f95e4)
Full Changelog: v2.7.1...v3.0.0
- api:
ResponseFunctionToolCallOutputItem.outputandResponseCustomToolCallOutput.outputnow returnstring | Array<ResponseInputText | ResponseInputImage | ResponseInputFile>instead ofstringonly. This may break existing callsites that assumeoutputis always a string.
- api: Support images and files for function call outputs in responses, BatchUsage (21901ef)
Full Changelog: v2.7.0...v2.7.1
- bugfix for setting JSON keys with special characters (f9ae028)
Full Changelog: v2.6.1...v2.7.0
- api: gpt-5-codex (b0eac3e)
Full Changelog: v2.6.0...v2.6.1
- api: fix mcp tool name (6de601a)
- use slices.Concat instead of sometimes modifying r.Options (7312ee7)
- api: openapi updates for conversations (4a7d204)
- bump minimum go version to 1.22 (8396ab5)
- do not install brew dependencies in ./scripts/bootstrap by default (d519b81)
- update more docs for 1.22 (1b0514d)
Full Changelog: v2.5.0...v2.6.0
- api: add reasoning_text (6ebf50d)
Full Changelog: v2.4.3...v2.5.0
- api: type updates for conversations, reasoning_effort and results for evals (3e68a60)
Full Changelog: v2.4.2...v2.4.3
- api: docs and spec refactoring (e67af66)
Full Changelog: v2.4.1...v2.4.2
- api: Minor docs and type updates for realtime (d92ea48)
Full Changelog: v2.4.0...v2.4.1
- api: fix realtime GA types (012b83e)
Full Changelog: v2.3.1...v2.4.0
- api: ship the RealtimeGA API shape (2b6c6db)
Full Changelog: v2.3.0...v2.3.1
- internal: unmarshal correctly when there are multiple discriminators (98596b2)
Full Changelog: v2.2.2...v2.3.0
- api: Add gpt-realtime models (3cf6a34)
Full Changelog: v2.2.1...v2.2.2
Full Changelog: v2.2.0...v2.2.1
- api: manual updates for ResponseInputAudio (8c0ebe5)
Full Changelog: v2.1.1...v2.2.0
- api: Add connectors support for MCP tool (35888bc)
- api: add web search filters (6f2c71d)
- api: adding support for /v1/conversations to the API (5b7c31b)
- api: realtime API updates (130fc8e)
- client: add support for verifying signatures on incoming webhooks (f7c8dbb)
Full Changelog: v2.1.0...v2.1.1
- api: accurately represent shape for verbosity on Chat Completions (f81197b)
Full Changelog: v2.0.2...v2.1.0
- api: add new text parameters, expiration options (323154c)
Full Changelog: v2.0.1...v2.0.2
Full Changelog: v2.0.0...v2.0.1
- client: fix verbosity parameter location in Responses (6e2e903)
Full Changelog: v1.12.0...v2.0.0
With the launch of custom tools in Chat Completions, function tools have been renamed to clarify the difference between the two.
ChatCompletionToolParam has become a union and is now named ChatCompletionToolUnionParam.
Older versions of the SDK used function tools: to migrate
- openai.ChatCompletionToolParam{
- Function: openai.FunctionDefinitionParam{
+ openai.ChatCompletionFunctionTool(
+ openai.FunctionDefinitionParam{
Name: "get_weather",
Description: openai.String("Get weather at the given location"),
Parameters: openai.FunctionParameters{ … },
+ },
+ )
- },- api: adds GPT-5 and new API features: platform.openai.com/docs/guides/gpt-5 (af46c88)
- api: manual updates (219f209)
- client: remove HTML escaping in JSON (aea5ebc)
- client: rename union helpers (645e881)
- client: support optional json html escaping (1d2336b)
- client: revert path param changes (9239f06)
- change readme warning and minimum version (1d0e22f)
- document breaking changes (afaa2b8)
- migrate examples (9c57dd7)
Full Changelog: v1.11.1...v1.12.0
- api: manual updates (16312ea)
- client: refactor streaming slightly to better future proof it (0b9cb85)
Full Changelog: v1.11.0...v1.11.1
- client: process custom base url ahead of time (cc1c23e)
- api: event shapes more accurate (2acd10d)
Full Changelog: v1.10.3...v1.11.0
- api: manual updates (97ed7fd)
Full Changelog: v1.10.2...v1.10.3
Full Changelog: v1.10.1...v1.10.2
- api: update realtime specs, build config (3d2afda)
Full Changelog: v1.10.0...v1.10.1
Full Changelog: v1.9.0...v1.10.0
- api: add file_url, fix event ID (cb33971)
Full Changelog: v1.8.3...v1.9.0
- client: expand max streaming buffer size (44390c8)
Full Changelog: v1.8.2...v1.8.3
- ci: only run for pushes and fork pull requests (d6aab99)
- internal: fix lint script for tests (9c0a745)
- lint tests (2bd38d2)
Full Changelog: v1.8.1...v1.8.2
- don't try to deserialize as json when ResponseBodyInto is []byte (74ad0f8)
- pagination: check if page data is empty in GetNextPage (c9becdc)
Full Changelog: v1.8.0...v1.8.1
- api: remove unsupported property (e22316a)
- docs: update README to include links to docs on Webhooks (7bb8f85)
Full Changelog: v1.7.0...v1.8.0
- api: webhook and deep research support (f6a7e7d)
- internal: add tests for breaking change detection (339522d)
Full Changelog: v1.6.0...v1.7.0
- api: make model and inputs not required to create response (19f0b76)
- api: update api shapes for usage and code interpreter (d24d42c)
- client: add escape hatch for null slice & maps (9c633d6)
- fix documentation of null map (8f3a134)
Full Changelog: v1.5.0...v1.6.0
- api: add reusable prompt IDs (280c698)
- api: manual updates (740f840)
- client: add debug log helper (5715c49)
- ci: enable for pull requests (9ed793a)
Full Changelog: v1.4.0...v1.5.0
- api: Add o3-pro model IDs (3bbd0b8)
Full Changelog: v1.3.0...v1.4.0
- client: allow overriding unions (27c6299)
- client: cast to raw message when converting to params (a3282b0)
Full Changelog: v1.2.1...v1.3.0
- api: add new realtime and audio models, realtime session options (8b8f62b)
Full Changelog: v1.2.0...v1.2.1
- make go mod tidy continue on error (48f41c2)
Full Changelog: v1.1.0...v1.2.0
- api: Config update for pakrym-stream-param (84d59d5)
- client: return binary content from
get /containers/{container_id}/files/{file_id}/content(f8c8de1)
Full Changelog: v1.0.0...v1.1.0
- api: add container endpoint (2bd777d)
- api: new API tools (e7e2123)
- api: new streaming helpers for background responses (422a0db)
Full Changelog: v0.1.0-beta.11...v1.0.0
- client: rename file array param variant
- api: improve naming and remove assistants
- accumulator: update casing (#401)
- api: improve naming and remove assistants (4c623b8)
- accumulator: update casing (#401) (d59453c)
- client: correctly set stream key for multipart (0ec68f0)
- client: don't panic on marshal with extra null field (9c15332)
- client: increase max stream buffer size (9456455)
- client: rename file array param variant (4cfcf86)
- client: use scanner for streaming (aa58806)
- docs: typo fix (#400) (bececf2)
- examples: migrate enum (#447) (814dd8b)
- examples: migrate to latest version (#444) (1c8754f)
- examples: remove beta assisstants examples (#445) (5891583)
- example: update fine-tuning (#450) (421e3c5)
Full Changelog: v0.1.0-beta.10...v0.1.0-beta.11
- client: clearer array variant names
- client: rename resp package
- client: improve core function names
- client: improve union variant names
- client: improve param subunions & deduplicate types
- api: add image sizes, reasoning encryption (0852fb3)
- api: add o3 and o4-mini model IDs (3fabca6)
- api: Add reinforcement fine-tuning api support (831a124)
- api: adding gpt-4.1 family of model IDs (1ef19d4)
- api: adding new image model support (bf17880)
- api: manual updates (11f5716)
- api: responses x eval api (183aaf7)
- api: Updating Assistants and Evals API schemas (47ca619)
- client: add dynamic streaming buffer to handle large lines (8e6aad6)
- client: add helper method to generate constant structs (ff82809)
- client: add support for endpoint-specific base URLs in python (072dce4)
- client: add support for reading base URL from environment variable (0d37268)
- client: clearer array variant names (a5d8b5d)
- client: experimental support for unmarshalling into param structs (5234875)
- client: improve param subunions & deduplicate types (8a78f37)
- client: rename resp package (4433516)
- client: support more time formats (ec171b2)
- fix lint (9c50a1e)
- client: clean up reader resources (710b92e)
- client: correctly update body in WithJSONSet (f2d7118)
- client: improve core function names (9f312a9)
- client: improve union variant names (a2c3de9)
- client: include path for type names in example code (69561c5)
- client: resolve issue with optional multipart files (910d173)
- client: time format encoding fix (ca17553)
- client: unmarshal responses properly (fc9fec3)
- handle empty bodies in WithJSONSet (8372464)
- pagination: handle errors when applying options (eebf84b)
- ci: add timeout thresholds for CI jobs (26b0dd7)
- ci: only use depot for staging repos (7682154)
- ci: run on more branches and use depot runners (d7badbc)
- docs: document pre-request options (4befa5a)
- docs: update respjson package name (9a00229)
- readme: improve formatting (a847e8d)
- utils: add internal resp to param utility (239c4e2)
- update documentation links to be more uniform (f5f0bb0)
Full Changelog: v0.1.0-beta.9...v0.1.0-beta.10
Full Changelog: v0.1.0-beta.8...v0.1.0-beta.9
Full Changelog: v0.1.0-beta.7...v0.1.0-beta.8
- api: Add evalapi to sdk (#360) (88977d1)
- api: manual updates (#363) (5d068e0)
- client: add escape hatch to omit required param fields (#354) (9690d6b)
- client: support custom http clients (#357) (b5a624f)
- docs: readme improvements (#356) (b2f8539)
- internal: fix examples (#361) (de398b4)
- internal: skip broken test (#362) (cccead9)
- tests: improve enum examples (#359) (e0b9739)
Full Changelog: v0.1.0-beta.6...v0.1.0-beta.7
Full Changelog: v0.1.0-beta.5...v0.1.0-beta.6
Full Changelog: v0.1.0-beta.4...v0.1.0-beta.5
- client: return error on bad custom url instead of panic (#341) (a06c5e6)
- client: support multipart encoding array formats (#342) (5993b28)
- client: unmarshal stream events into fresh memory (#340) (52c3e08)
Full Changelog: v0.1.0-beta.3...v0.1.0-beta.4
- docs: update file uploads in README (#333) (471c452)
- internal: codegen related update (#335) (48422dc)
- Remove deprecated/unused remote spec feature (c5077a1)
Full Changelog: v0.1.0-beta.2...v0.1.0-beta.3
- client: add enums (#327)
- api: add
get /chat/completionsendpoint (e8ed116) - api: add
get /responses/{response_id}/input_itemsendpoint (8870c26)
- add hash of OpenAPI spec/config inputs to .stats.yml (104b786)
- add request options to client tests (#321) (f5239ce)
- api: updates to supported Voice IDs (#325) (477727a)
- docs: improve security documentation (#319) (0271053)
- fix typos (#324) (dba15f7)
Full Changelog: v0.1.0-beta.1...v0.1.0-beta.2
Full Changelog: v0.1.0-alpha.67...v0.1.0-beta.1
Full Changelog: v0.1.0-alpha.66...v0.1.0-alpha.67
- api: migrate to v2
- api: migrate to v2 (9377508)
- api: new models for TTS, STT, + new audio features for Realtime (#298) (48fa064)
- internal: bugfix (0d8c1f4)