Revision history for OpenAPI-Client-OpenAI
0.28 2026-05-23
- Documentation cleanup: rewrote the SYNOPSIS code example
(was missing a comma after `prompt`, used the deprecated
createCompletion endpoint, and referenced Data::Dumper
without `use`-ing it). It now demonstrates the snake_case
create_chat_completion form against gpt-3.5-turbo.
- Regenerated README.md from POD so it reflects the snake_case
aliases being first-class and drops the now-removed
DEPRECATED METHODS section.
0.27 2026-05-03
- Added devel/browse-docs to see the new docs prior to release
- Snake_case aliases for every operation: every operationId is
callable in both its original style (e.g. createChatCompletion)
and snake_case (e.g. create_chat_completion). Both forms are
first-class; the deprecation warning on the legacy six aliases
is removed.
- Documentation generator rewrite: per-path POD is now produced by
a modular generator under devel/lib/OpenAPI/Client/OpenAI/DocGen*.
Notable fixes:
* snake_case Perl method in every code example (previously,
kebab-case operationIds produced uncompilable examples).
* Methods.pod is now auto-generated (alphabetical, ~5 lines
per method), replacing a stale hand-maintained file.
* JSON examples are normalized (no more nested stringified
JSON) and oneOf/anyOf variants are scored by informativeness
instead of YAML order.
* Method-level x-oaiMeta.examples are now used in addition
to schema-level examples.
* Markdown horizontal rules no longer leak 80-character `='
artifacts into the POD.
* Pod::Checker passes with zero errors and zero warnings
across all generated POD.
- Build dependencies dropped: Markdown::Pod, Template, Text::Wrap,
File::Slurp, Clone, Feature::Compat::Try.
0.26 2026-05-23
- Image Model Retirement: `dall-e-2` and `dall-e-3` were removed from
the OpenAI API on 2026-05-12. The `model` parameter to `createImage`
no longer defaults to a working value and must be passed explicitly.
Use `gpt-image-1.5`, `gpt-image-1`, or `gpt-image-1-mini` instead.
The `t/create_image.t` test has been updated accordingly.
- ChatKit API: New top-level `/chatkit` endpoints for sessions and
threads, including item management and session cancellation.
- Skills API: New `/skills` endpoints for creating, listing, and
versioning reusable skill artifacts, including per-version content.
- Videos API: New `/videos` endpoints covering video generation,
editing, extension, remixing, content retrieval, and reusable video
characters.
- Realtime API Expansion: New `/realtime/calls/{call_id}` endpoints
for SIP-style call control (`accept`, `hangup`, `refer`, `reject`),
plus new `/realtime/client_secrets` and
`/realtime/translations/client_secrets` endpoints.
- Organization Administration: New endpoints for spend alerts (at
both organization and project scope), data retention policies,
groups, roles, hosted tool permissions, and model permissions.
- Usage Reporting: New `/organization/usage/file_search_calls` and
`/organization/usage/web_search_calls` endpoints.
- Responses API: Added `/responses/compact` and
`/responses/input_tokens` endpoints.
- Build Tooling: `devel/rebuild` now runs `make manifest` after
regenerating documentation so newly-generated `.pod` files are
automatically added to MANIFEST.
0.25 2026-01-02
- Model Updates: Updated examples and internal constants to reflect the
latest `gpt-5.2` series models.
- Evals API Introduction: Added a comprehensive set of endpoints for
managing model evaluations, including creating evaluations, executing
evaluation runs, and retrieving detailed output items.
- Audio API Enhancements:
- Introduced support for speaker diarization in transcriptions via the
new `diarized_json` response format.
- Added new endpoints for managing Voice Consents and creating custom
Voices.
- Conversations API Updates:
- Restructured conversation item management under the
`/conversations/{conversation_id}/items` path.
- Added the `include` parameter to allow fetching enhanced metadata
such as tool outputs and reasoning tokens.
- Realtime API (Beta) Updates: Expanded server events to include support
for Model Context Protocol (MCP) tool calls and audio transcript
deltas.
- Assistant API Enhancements:
- Added support for new tools, including `web_search_preview` and
`computer_use_preview`.
- Introduced the `reasoning_effort` parameter and prompt cache
retention policies for assistants.
- Batch API Expansion: Added support for including Moderations
(`/v1/moderations`) in batch requests.
- Containers API: Added `memory_limit` configuration support for
container objects.
- Documentation: Updated all code examples to prioritize the use of
the `OPENAI_API_KEY` environment variable.
0.24 2025-08-22
- Model Updates: Added support for new `gpt-5` series models, including
`gpt-5`, `gpt-5-mini`, and `gpt-5-nano`, across various API endpoints like
Assistants and Chat Completions.
- Conversations API Introduction: A new top-level `/conversations` endpoint has
been added to create, retrieve, update, and delete conversations and their
items. This provides a new way to manage and persist conversation state. The
previous `/embeddings` endpoint has been removed to accommodate this new
feature.
- Images API Streaming: The Images API now supports streaming for both image
generation (`/images/generations`) and editing (`/images/edits`). New
parameters such as `input_fidelity` and `partial_images` have also been
introduced for more control over image results.
- Documentation and Code Examples: Replaced all Kotlin code examples with Java
examples across the entire API specification. Endpoint summaries have been
standardized for conciseness, with longer explanations moved into the
`description` field. Additionally, internal documentation links have been
updated to absolute URLs.
- File API Enhancements: The organizational file storage limit has been
increased to 1 TB. A new `expires_after` parameter allows setting an expiration
policy upon file creation.
- Responses API Updates: A new `conversation` parameter has been added to
manage conversation state more effectively. The `user` parameter is now
deprecated in favor of `safety_identifier` and `prompt_cache_key` for better
user identification and caching. A `verbosity` parameter has also been added to
control response conciseness.
- Other API Updates: The Batch API now includes an `output_expires_after`
parameter to manage the lifecycle of output files. Many schema definitions have
been updated from `oneOf` to the more flexible `anyOf`.
0.23 2025-07-03
- Version number bump to replace bad PAUSE upload. No functional change.
0.22 2025-07-03
- Added new `git-release` script to automate the release process.
- Audio API Updates: The audio text-to-speech model is updated to
gpt-4o-mini-tts, and the audio transcription model is updated to
gpt-4o-transcribe, with the addition of a streaming option for the
transcription api.
- Chat Completions Enhancements: Introduces new features for the Chat
Completions API, including the ability to list, retrieve, update, and delete
chat completions. Support for metadata filtering is added, and the
documentation clarifies parameter support across different models.
- Realtime API Expansion: Adds a new endpoint for creating realtime
transcription sessions and incorporates C# examples for both audio generation
and transcription.
- Responses API Improvements: Significant updates to the Responses
API, with a focus on enhanced tool usage, including web search and file search,
and the ability to stream responses. The API now also supports including image
URLs in the input and provides detailed usage information.
- Fine-tuning API Updates: Introduces new endpoints for managing
fine-tuning checkpoints and their permissions, enabling organization owners to
share and control access to fine-tuned models. Metadata is now supported for
fine-tuning jobs.
- Other Updates: Minor documentation improvements and clarifications
across various API endpoints and parameters, including updates to model
descriptions and available tools.
0.21 2025-04-11
- Added examples to the docs, along with allowed models for each path.
0.20 2025-04-06
- Complete revamped documentation. Much easier to read than the
near-useless Schema.pod that we used to have.
0.16 2025-04-06
- Audio API Updates: The audio text-to-speech model is updated to
gpt-4o-mini-tts, and the audio transcription model is updated to
gpt-4o-transcribe, with the addition of a streaming option for the
transcription api.
- Chat Completions Enhancements: Introduces new features for the Chat
Completions API, including the ability to list, retrieve, update, and delete
chat completions. Support for metadata filtering is added, and the
documentation clarifies parameter support across different models.
- Realtime API Expansion: Adds a new endpoint for creating realtime
transcription sessions and incorporates C# examples for both audio generation
and transcription.
- Responses API Improvements: Significant updates to the Responses
API, with a focus on enhanced tool usage, including web search and file search,
and the ability to stream responses. The API now also supports including image
URLs in the input and provides detailed usage information.
- Fine-tuning API Updates: Introduces new endpoints for managing
fine-tuning checkpoints and their permissions, enabling organization owners to
share and control access to fine-tuned models. Metadata is now supported for
fine-tuning jobs.
- Other Updates: Minor documentation improvements and clarifications
across various API endpoints and parameters, including updates to model
descriptions and available tools.
0.15 2025-02-28
- Update to latest version of OpenAI API spec.
Even though the 4.5-preview has been released
(https://platform.openai.com/docs/models#gpt-4-5), it's not listed
in their openapi.yaml document. The changes are largely around the
documentation in the API spec.
0.14 2025-02-05
- Update to latest version of OpenAI API spec. Amongst many changes,
this includes the new o3-mini model.
0.13 2025-01-02
- Happy New Year!
- Update assistance header to v2 https://platform.openai.com/docs/assistants/migration
- Update to latest version of OpenAI API spec
- Cleaned up main class imports
0.12 2024-11-18
- Added examples/get-embeddings.pl to the MANIFEST
0.11 2024-11-18
- Update to the latest OpenAPI spec
- Add example of fetching embeddings from the API
0.10 2024-10-05
- Update MANIFEST to include missing example files
0.09 2024-10-05
- Update OpenAI spec
- Include examples/describe_image.pl
0.08 2024-09-19 12:24 (CEST)
- Bump OpenAI spec to version 2.3.0
0.07 2024-07-07 00:00 (CEST)
- Added a missing dependencies in Makefile.PL
- Add documentation to explain the deprecated methods.
0.06 2024-05-29 08:22 (CEST)
- Build bugfix: File::ShareDir::Install needed to be in
CONFIGURE_REQUIRES in the Makefile.PL. Thanks to Slaven Rezić
for the bug report
- POD now assumes UTF-8
0.05 2024-05-29 07:16 (CEST)
- Added `devel/rebuild` script to make it easier to rebuild the
project.
- Common test behavior abstracted into t/lib/OpenAITests.pm
- Bugfix: rebuilding docs should only list Perl files for examples.
0.04 2024-05-27 16:06 (CEST)
- Indent HEREDOCS so metacpan doesn't think they're POD
- Fix bad link in the POD
- Update POD to show where to find examples
0.03 2024-05-27
- Update manifest to ensure files are properly uploaded
0.02 2024-05-15
- Updated to version 2.0.0 of openapi.yaml
0.01 2023-04-11
First version, released on an unsuspecting world.