Skip to main content
The Client class is the main entry point for making synchronous requests to the Gemini Developer API or Vertex AI API. It provides access to all SDK features including models, files, caches, tunings, and more.

​
Constructor

Create a new client instance for either the Gemini API or Vertex AI API.
​
vertexai
bool
default:"False"
Indicates whether the client should use the Vertex AI API endpoints. Defaults to False (uses Gemini Developer API endpoints).
​
api_key
str
The API key to use for authentication. Applies to the Gemini Developer API only. Can also be set via the GOOGLE_API_KEY environment variable.
​
credentials
google.auth.credentials.Credentials
The credentials to use for authentication when calling the Vertex AI APIs. Credentials can be obtained from environment variables and default credentials. For more information, see Set up Application Default Credentials. Applies to the Vertex AI API only.
​
project
str
The Google Cloud project ID to use for quota. Can be obtained from environment variables (e.g., GOOGLE_CLOUD_PROJECT). Applies to the Vertex AI API only. Find your project ID.
​
location
str
The location to send API requests to (e.g., us-central1). Can be obtained from environment variables (e.g., GOOGLE_CLOUD_LOCATION). Applies to the Vertex AI API only.
​
debug_config
DebugConfig
Configuration settings that control network behavior of the client. This is typically used when running test code.
​
http_options
types.HttpOptions | dict
HTTP options to use for the client. These options will be applied to all requests made by the client.See HttpOptions for available fields.

​
Usage Examples

​
Gemini Developer API

from google import genai

client = genai.Client(api_key='my-api-key')

​
Vertex AI API

from google import genai

client = genai.Client(
    vertexai=True,
    project='my-project-id',
    location='us-central1'
)

​
With HTTP Options

from google import genai
from google.genai import types

client = genai.Client(
    api_key='my-api-key',
    http_options=types.HttpOptions(
        api_version='v1',
        timeout=30000,  # 30 seconds in milliseconds
        headers={'Custom-Header': 'value'}
    )
)

​
Using Context Manager

from google import genai

with genai.Client(api_key='my-api-key') as client:
    response = client.models.generate_content(
        model='gemini-2.0-flash',
        contents='Hello World'
    )
    print(response.text)
# Client is automatically closed when exiting the context

​
Properties

​
models

​
models
Models
Access to the Models API for content generation, embeddings, and model management.See generate_content for details.
response = client.models.generate_content(
    model='gemini-2.0-flash',
    contents='Tell me a story'
)

​
aio

​
aio
AsyncClient
Access to the asynchronous client for non-blocking operations.See AsyncClient for details.
async_client = client.aio
response = await async_client.models.generate_content(
    model='gemini-2.0-flash',
    contents='Hello async world'
)

​
chats

​
chats
Chats
Access to multi-turn conversation functionality.See chats.create for details.
chat = client.chats.create(model='gemini-2.0-flash')
response = chat.send_message('Hello!')

​
files

​
files
Files
Access to the Files API for uploading and managing media files.See files.upload for details.
file = client.files.upload(path='image.jpg')
print(file.name)

​
caches

​
caches
Caches
Access to the Caches API for context caching.See caches.create for details.
cache = client.caches.create(
    model='gemini-2.0-flash',
    contents='Long document...'
)

​
file_search_stores

​
file_search_stores
FileSearchStores
Access to the File Search Stores API for semantic search.

​
batches

​
batches
Batches
Access to the Batches API for batch processing.See batches.create for details.

​
tunings

​
tunings
Tunings
Access to the Tunings API for model fine-tuning.See tunings.tune for details.

​
auth_tokens

​
auth_tokens
Tokens
Access to authentication token management.

​
operations

​
operations
Operations
Access to long-running operations management.

​
interactions

​
interactions
InteractionsResource
Access to the experimental Interactions API for live, streaming interactions.
This API is experimental and may change in future versions.

​
vertexai

​
vertexai
bool
Returns whether the client is using the Vertex AI API.
if client.vertexai:
    print("Using Vertex AI API")
else:
    print("Using Gemini Developer API")

​
Methods

​
close()

Closes the synchronous client explicitly and releases resources.
This method does not close the async client. Use client.aio.aclose() or the async context manager to close the async client.
from google.genai import Client

client = Client(
    vertexai=True,
    project='my-project-id',
    location='us-central1'
)

response_1 = client.models.generate_content(
    model='gemini-2.0-flash',
    contents='Hello World'
)

response_2 = client.models.generate_content(
    model='gemini-2.0-flash',
    contents='Goodbye World'
)

# Close the client to release resources
client.close()

​
Context Manager Methods

The Client supports the context manager protocol for automatic resource cleanup.

​
__enter__()

Enters the runtime context and returns the client.

​
__exit__(exc_type, exc_value, traceback)

Exits the runtime context and closes the client. 
with genai.Client(api_key='my-api-key') as client:
    # Use the client
    response = client.models.generate_content(
        model='gemini-2.0-flash',
        contents='Hello'
    )
# Client is automatically closed here

​
HttpOptions

HTTP configuration options for customizing client behavior.
​
base_url
str
The base URL for the AI platform service endpoint.
​
api_version
str
Specifies the version of the API to use.
​
headers
dict[str, str]
Additional HTTP headers to be sent with each request.
​
timeout
int
Timeout for requests in milliseconds.
​
client_args
dict[str, Any]
Arguments passed to the HTTP client.
​
async_client_args
dict[str, Any]
Arguments passed to the async HTTP client.
​
extra_body
dict[str, Any]
Extra parameters to add to the request body. The structure must match the backend API’s request structure:
​
retry_options
HttpRetryOptions
HTTP retry options for requests. See HttpRetryOptions.
​
httpx_client
httpx.Client
A custom httpx client to be used for requests.
​
httpx_async_client
httpx.AsyncClient
A custom httpx async client to be used for requests.
​
aiohttp_client
aiohttp.ClientSession
A custom aiohttp client session to be used for requests.

​
HttpRetryOptions

Configuration options for HTTP request retries.
​
attempts
int
default:"5"
Maximum number of attempts, including the original request. If 0 or 1, it means no retries.
​
initial_delay
float
default:"1.0"
Initial delay before the first retry, in seconds.
​
max_delay
float
default:"60.0"
Maximum delay between retries, in seconds.
​
exp_base
float
default:"2.0"
Multiplier by which the delay increases after each attempt.
​
jitter
float
default:"1.0"
Randomness factor for the delay.
​
http_status_codes
list[int]
List of HTTP status codes that should trigger a retry. If not specified, a default set of retryable codes (408, 429, and 5xx) may be used.
from google import genai
from google.genai import types

client = genai.Client(
    api_key='my-api-key',
    http_options=types.HttpOptions(
        retry_options=types.HttpRetryOptions(
            attempts=3,
            initial_delay=0.5,
            max_delay=10.0
        )
    )
)

​
See Also

Build docs developers (and LLMs) love

Get started for freeTalk to us