openai-openapi
openai-openapi copied to clipboard
openai
OpenAPI specification for the OpenAI API
I am giving gpt4o a paged document, that is in a markdown format with page separators like those: ```markdown {PAGE 1}------------------------------------------------  **PLIEGO DE CLÁUSULAS ADMINISTRATIVAS PARTICULARES QUE HA DE...
In the property `json_schema` of the `ResponseFormatJsonSchema` it mentions a required `type` property: https://github.com/openai/openai-openapi/blob/master/openapi.yaml#L19896. However, this `type` property is actually not part of the `ResponseFormatJsonSchema.json_schema` object properties: https://github.com/openai/openai-openapi/blob/e0cb2d721753e13e69e918465795d6e9f87ab15a/openapi.yaml#L19865-L19897 So either...
in many of the schemas' `model` property. For example `CreateChatCompletionResponse, CreateRunRequest, etc.`
the transcription response should also have `text/plain` content type for art, vtt, and text response_formats. the last 3 lines should be added. ```yaml /audio/transcriptions: post: operationId: createTranscription tags: - Audio...
Specification: https://www.asyncapi.com/en Example: https://github.com/AssemblyAI/assemblyai-api-spec/blob/main/asyncapi.yml
Hello. While trying to create a clone of the API in Azure API Management using the version on master, I am getting validation errors of the file. https://raw.githubusercontent.com/openai/openai-openapi/refs/heads/master/openapi.yaml 
Completion Usage is now returning prompt_tokens_details as well which is not available in openapi specs "usage": { "prompt_tokens": 38, "completion_tokens": 16, "total_tokens": 54, "prompt_tokens_details": { "cached_tokens": 0 }, "completion_tokens_details": {...
According to the Typescript library, `tool_calls` is defined as `tool_calls?: Array` and the Python library allows `tool_calls` to be null - however, this OpenAPI spec defines `tool_calls` on `ChatCompletionRequestAssistantMessage` as...
With the recent release of [prompt caching](https://platform.openai.com/docs/guides/prompt-caching), there are new properties for 'cached_tokens'. ``` "usage": { "prompt_tokens": 2006, "completion_tokens": 300, "total_tokens": 2306, "prompt_tokens_details": { "cached_tokens": 1920 }, "completion_tokens_details": { "reasoning_tokens":...
