openai-openapi
openai-openapi copied to clipboard
openai
OpenAPI specification for the OpenAI API
`createSpeech` functionality specifies `response.content` as `application/octet-stream` but every request returns more specific mime type, such as `audio/mpeg` or `audio/aac`.
openapi-generator-cli generate -i src/openapi.yaml -g typescript-fetch -o src/openai-client Exception in thread "main" org.openapitools.codegen.SpecValidationException: There were issues with the specification. The option can be disabled via validateSpec (Maven/Gradle) or --skip-validate-spec (CLI)....
Why were the `x-oaiMeta/group` keys removed from the spec? It allowed a nice categorization of endpoints in generated client libraries.
See https://platform.openai.com/docs/guides/text-to-speech/supported-output-formats
This makes it easier for us to generate a named type for this schema.
Related: https://github.com/openai/openai-python/issues/1089 This proposed change would help to add clarity for python/node SDKs where it's unclear that for the Azure API, the deployment name of the model should be passed...
it is present in the URL, which is where it is already required.
An example request of speech creation in Python that still uses an old method [documentation](https://platform.openai.com/docs/api-reference/audio/createSpeech) I updated it to client variable.
The API reference mentions `gpt-4-visual-preview` instead of the correct model name for `gpt-4` with vision capabilities i.e. `gpt-4-vision-preview`
