Extract typed JSON from text with Gemini 3.8 Flash and Pydantic

What you get

Turn messy text into validated, typed JSON with Gemini 3.8 Flash by passing a Pydantic schema as the response format.

Prerequisites

A Gemini API key in GEMINI_API_KEY
Python with google-genai and pydantic

Steps

Install the SDK and set your key.

Shell

pip install -U google-genai pydantic
export GEMINI_API_KEY="YOUR_API_KEY"

Save as extract.py: define the schema and call the Interactions API with response_format.

Python

from google import genai
from pydantic import BaseModel, Field
from typing import List, Optional

class Ingredient(BaseModel):
    name: str = Field(description="Name of the ingredient.")
    quantity: str = Field(description="Quantity of the ingredient, including units.")

class Recipe(BaseModel):
    recipe_name: str = Field(description="The name of the recipe.")
    prep_time_minutes: Optional[int] = Field(description="Optional time in minutes to prepare the recipe.")
    ingredients: List[Ingredient]
    instructions: List[str]

client = genai.Client()

text = "Cookies: 2 1/4 cups flour, 1 tsp baking soda, 2 eggs. Preheat to 375F, mix, bake 9-11 minutes."

interaction = client.interactions.create(
    model="gemini-3.8-flash",
    input=f"Please extract the recipe from the following text.\n{text}",
    response_format={
        "type": "text",
        "mime_type": "application/json",
        "schema": Recipe.model_json_schema()
    },
)

recipe = Recipe.model_validate_json(interaction.output_text)
print(recipe)

Test it

Shell

python extract.py

You should get a Recipe object with recipe_name, a list of ingredients and instructions. Swap in your own Pydantic model for invoices, leads or support tickets.