support for JSON output mode in ChatGoogleGenerativeAI

[cookiejest]

Checked other resources

• [X] I added a very descriptive title to this issue.
• [X] I searched the LangChain documentation with the integrated search.
• [X] I used the GitHub search to find a similar question and didn't find it.
• [X] I am sure that this is a bug in LangChain rather than my code.
• [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).

Example Code

The underlying googlegemini supports JSON mode and response type with gemini. Is there a way to update the documentation with an example of this?

` response_schema = { "type": "ARRAY", "items": { "type": "OBJECT", "properties": { "recipe_name": { "type": "STRING", }, }, "required": ["recipe_name"], }, }

response = model.generate_content( "List a few popular cookie recipes", generation_config=GenerationConfig( response_mime_type="application/json", response_schema=response_schema ), ) `

Error Message and Stack Trace (if applicable)

No response

Description

im trying to use gemini pro 1.5 with forced JSON output as documented here: https://github.com/GoogleCloudPlatform/generative-ai/blob/main/gemini/controlled-generation/intro_controlled_generation.ipynb

System Info

langchain==0.2.12 langchain_core==0.2.29 langchain_google_genai==1.0.8 google-generativeai==0.7.2

[DenizK00]

Until the JSON output support is done, I think using pydantic model with JsonOutputParser on the unstructured/text output has a good performance. Even though it requires an additional call, validity-wise it's good.