response_schema not work for generate_content_stream

[hjlarry]

Thanks for stopping by to let us know something could be better!

PLEASE READ: If you have a support contract with Google, please create an issue in the support console instead of filing on GitHub. This will ensure a timely response.

Is this a client library issue or a product issue? We will only be able to assist with issues that pertain to the behaviors of this library. If the issue you're experiencing is due to the behavior of the product itself, please visit the Support page to reach the most relevant engineers.

If the support paths suggested above still do not result in a resolution, please provide the following details.

Environment details

• Programming language: Python
• OS: windows local develop env
• Language runtime version: python 3.12
• Package version: google-genai-0.2.2

Steps to reproduce

I follow the tutorial of json_schema, but use the stream mode, will raise error.

This is my code:

response = client.models.generate_content_stream(
    model='gemini-2.0-flash-exp',
    contents='Give me information of the United States.',
    config={
        'response_mime_type': 'application/json',
        'response_schema': {
            'properties': {
                'name': {'type': 'STRING'},
                'population': {'type': 'INTEGER'},
                'capital': {'type': 'STRING'},
                'continent': {'type': 'STRING'},
                'gdp': {'type': 'INTEGER'},
                'official_language': {'type': 'STRING'},
                'total_area_sq_mi': {'type': 'INTEGER'},
            },
            'type': 'OBJECT',
        },
    },
)

for r in response:
    parts = r.candidates[0].content.parts
    for p in parts:
        print(p.text)

This is the error:

Traceback (most recent call last):
  File "C:\Users\hejl\PycharmProjects\dify\api\.venv\Lib\site-packages\google\genai\models.py", line 3723, in generate_content_stream
    return_value = types.GenerateContentResponse._from_response(
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\hejl\PycharmProjects\dify\api\.venv\Lib\site-packages\google\genai\types.py", line 2464, in _from_response
    result.parsed = json.loads(result.text)
                    ^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\hejl\.pyenv\pyenv-win\versions\3.12.0\Lib\json\__init__.py", line 346, in loads
    return _default_decoder.decode(s)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\hejl\.pyenv\pyenv-win\versions\3.12.0\Lib\json\decoder.py", line 337, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\hejl\.pyenv\pyenv-win\versions\3.12.0\Lib\json\decoder.py", line 353, in raw_decode
    obj, end = self.scan_once(s, idx)
               ^^^^^^^^^^^^^^^^^^^^^^
json.decoder.JSONDecodeError: Expecting property name enclosed in double quotes: line 1 column 2 (char 1)

if I request with client.models.generate_content it works, but I need the stream, Thanks!

[Ark-kun]

Hello. Thank you for the issue report.

What are your expectation for streaming a JSON object?

When you receive a JSON data chunk like {"prop1": 1, "pro, what are you going to do with this data?

[hjlarry]

Thanks for reply.

I don't require parts of JSON data, but the dify platform use stream mode for all user chat requests, which works with the old gemini sdk. However, it will raise error with the new sdk.