Implementation:Googleapis Python genai GenerateContentConfig Setup

Knowledge Sources	googleapis/python-genai Google Gen AI Python SDK
Domains	NLP, Decoding_Strategies
Last Updated	2026-02-15 00:00 GMT

Overview

Concrete tool for specifying generation parameters and sampling strategies provided by the google-genai types module.

Description

GenerateContentConfig is a Pydantic model that holds all optional configuration for content generation requests. It controls sampling parameters (temperature, top_p, top_k), output format (response_mime_type, response_schema), safety settings, system instructions, tool configurations, and caching references. It is passed as the config parameter to models.generate_content and models.generate_content_stream.

Usage

Create a GenerateContentConfig when you need to customize generation behavior. Pass it as a dict or Pydantic model to the config parameter. For default behavior, omit the config entirely. Common use cases include setting a system instruction, adjusting temperature, or enabling JSON output mode.

Code Reference

Source Location

• Repository: googleapis/python-genai
• File: google/genai/types.py
• Lines: L5077

Signature

class GenerateContentConfig(_common.BaseModel):
    """Optional model configuration parameters."""
    http_options: Optional[HttpOptions] = None
    system_instruction: Optional[ContentUnion] = None
    temperature: Optional[float] = None
    top_p: Optional[float] = None
    top_k: Optional[float] = None
    candidate_count: Optional[int] = None
    max_output_tokens: Optional[int] = None
    stop_sequences: Optional[list[str]] = None
    response_mime_type: Optional[str] = None
    response_schema: Optional[SchemaUnion] = None
    response_modalities: Optional[list[str]] = None
    safety_settings: Optional[list[SafetySetting]] = None
    tools: Optional[list[ToolUnion]] = None
    tool_config: Optional[ToolConfig] = None
    cached_content: Optional[str] = None
    response_logprobs: Optional[bool] = None
    logprobs: Optional[int] = None
    presence_penalty: Optional[float] = None
    frequency_penalty: Optional[float] = None
    seed: Optional[int] = None
    automatic_function_calling_config: Optional[AutomaticFunctionCallingConfig] = None

Import

from google.genai import types

I/O Contract

Inputs

Name	Type	Required	Description
temperature	Optional[float]	No	Sampling temperature (0.0 = deterministic, higher = more random)
top_p	Optional[float]	No	Nucleus sampling threshold
top_k	Optional[float]	No	Top-k token count
candidate_count	Optional[int]	No	Number of response candidates to generate
max_output_tokens	Optional[int]	No	Maximum tokens in the response
system_instruction	Optional[ContentUnion]	No	System-level behavioral instruction
response_mime_type	Optional[str]	No	Output MIME type (e.g., 'application/json')
response_schema	Optional[SchemaUnion]	No	JSON schema for structured output
tools	Optional[list[ToolUnion]]	No	Tool/function definitions
cached_content	Optional[str]	No	Cache resource name for context caching
safety_settings	Optional[list[SafetySetting]]	No	Safety filter configurations
seed	Optional[int]	No	Random seed for reproducibility

Outputs

Name	Type	Description
GenerateContentConfig	GenerateContentConfig	Configuration object to pass as config to generate_content

Usage Examples

Basic Configuration

from google import genai
from google.genai import types

client = genai.Client(api_key="YOUR_API_KEY")

response = client.models.generate_content(
    model="gemini-2.0-flash",
    contents="Write a haiku about programming.",
    config=types.GenerateContentConfig(
        temperature=0.9,
        max_output_tokens=100,
        system_instruction="You are a creative poet.",
    ),
)
print(response.text)

JSON Output Mode

from google import genai
from google.genai import types

client = genai.Client(api_key="YOUR_API_KEY")

response = client.models.generate_content(
    model="gemini-2.0-flash",
    contents="List 3 programming languages with their creators.",
    config=types.GenerateContentConfig(
        response_mime_type="application/json",
        temperature=0.0,
    ),
)
print(response.text)  # Returns JSON string

Related Pages

Implements Principle

• Principle:Googleapis_Python_genai_Generation_Configuration