Unable to create json schema assistants in async

[jmesich]

Confirm this is an issue with the Python library and not an underlying OpenAI API

• This is an issue with the Python library

Describe the bug

When I try to create a json schema assistant, I get this error
"error": "Error code: 400 - {'error': {'message': \"Missing required parameter: 'response_format.json_schema.schema'. You provided 'schema_', did you mean to provide 'schema'?\", 'type': 'invalid_request_error', 'param': 'response_format.json_schema.schema', 'code': 'missing_required_parameter'}}"

When I go into the response_format_json_schema.py file and update JSONSchema.schema_ to JSONSchema.schema, the problem disappears.

To Reproduce

Call openai.AsyncOpenAI().beta.assistants.create with a response_format parameter

Code snippets

await openai.AsyncOpenAI().beta.assistants.create(
            model=assistant.model,
            description=assistant.description,
            instructions=assistant.instructions,
            name=assistant.name,
            response_format=assistant.response_format,
            temperature=assistant.temperature,
            tool_resources=assistant.tool_resources,
            tools=assistant.tools,
            top_p=assistant.top_p,
)

OS

MacOS Sonoma 14.6.1

Python version

Python v3.12.2

Library version

openai v1.54.4

[Dhruv-net]

To address this issue, you can modify the library's source code to ensure that schema_ is correctly serialized as schema in the API request. This involves updating the serialization logic within the response_format_json_schema.py file. By changing the attribute from schema_ to schema, the API will recognize the parameter, resolving the error.

[jmesich]

Hi @Dhruv-net, thank you for your response but I have already mentioned that fix in my description above and I do not wish to fork the repo for my deployments!