Os modelos do Gemini podem ser acessados usando as bibliotecas da OpenAI (Python, TypeScript e JavaScript) com a API REST. Somente a autenticação do Google Cloud é compatível com a biblioteca da OpenAI na Gemini Enterprise Agent Platform. Se você ainda não usa as bibliotecas OpenAI, chame a API Gemini diretamente. Se você estiver usando as bibliotecas da OpenAI e quiser migrar para os SDKs da Agent Platform, consulte Migrar do SDK da OpenAI para o SDK de IA generativa do Google.
Python
import openai
from google.auth import default
import google.auth.transport.requests
# TODO(developer): Update and un-comment below lines
# project_id = "PROJECT_ID"
# location = "global"
# Programmatically get an access token
credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
credentials.refresh(google.auth.transport.requests.Request())
# OpenAI Client
client = openai.OpenAI(
base_url=f"https://aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/openapi",
api_key=credentials.token
)
response = client.chat.completions.create(
model="google/gemini-3.5-flash",
messages=[
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Explain to me how AI works"}
]
)
print(response.choices[0].message)
Observe as seguintes diferenças em relação à configuração padrão do cliente da OpenAI:
api_key=credentials.token: usa um token de acesso OAuth recebido do Google Cloud Application Default Credentials.base_url: direciona a biblioteca da OpenAI para enviar solicitações ao endpoint compatível com a OpenAI da plataforma de agentes em vez do URL padrão.model="google/gemini-3.5-flash": seleciona um modelo do Gemini compatível hospedado na plataforma de agentes.
Pensando
Os modelos do Gemini 3 geram um processo de raciocínio interno antes de
retornar uma resposta, melhorando o desempenho em tarefas complexas de várias etapas. Na
API Gemini, o
parâmetro thinking_level
controla a profundidade do raciocínio em níveis discretos (MINIMAL, LOW, MEDIUM e
HIGH).
Na API Chat Completions compatível com a OpenAI, você controla a profundidade do raciocínio com
o parâmetro reasoning_effort ("minimal", "low", "medium" ou
"high"), que é mapeado para a configuração thinking_level correspondente para
modelos do Gemini 3.
Se você omitir reasoning_effort, o modelo vai usar a configuração de raciocínio padrão.
Para ter controle direto sobre os parâmetros de configuração de pensamento da
API compatível com a OpenAI, use
extra_body.google.thinking_config.
Python
import openai
from google.auth import default
import google.auth.transport.requests
# TODO(developer): Update and un-comment below lines
# project_id = "PROJECT_ID"
# location = "global"
# Programmatically get an access token
credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
credentials.refresh(google.auth.transport.requests.Request())
# OpenAI Client
client = openai.OpenAI(
base_url=f"https://aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/openapi",
api_key=credentials.token
)
response = client.chat.completions.create(
model="google/gemini-3.5-flash",
reasoning_effort="low",
messages=[
{"role": "system", "content": "You are a helpful assistant."},
{
"role": "user",
"content": "Explain to me how AI works"
}
]
)
print(response.choices[0].message)
Streaming
A API Gemini é compatível com respostas de streaming.
Python
import openai
from google.auth import default
import google.auth.transport.requests
# TODO(developer): Update and un-comment below lines
# project_id = "PROJECT_ID"
# location = "global"
credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
credentials.refresh(google.auth.transport.requests.Request())
client = openai.OpenAI(
base_url=f"https://aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/openapi",
api_key=credentials.token
)
response = client.chat.completions.create(
model="google/gemini-3.5-flash",
messages=[
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Hello!"}
],
stream=True
)
for chunk in response:
print(chunk.choices[0].delta)
Chamadas de função
A chamada de função facilita a obtenção de saídas de dados estruturados de modelos generativos e é compatível com a API Gemini.
Python
import openai
from google.auth import default
import google.auth.transport.requests
# TODO(developer): Update and un-comment below lines
# project_id = "PROJECT_ID"
# location = "global"
credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
credentials.refresh(google.auth.transport.requests.Request())
client = openai.OpenAI(
base_url=f"https://aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/openapi",
api_key=credentials.token
)
tools = [
{
"type": "function",
"function": {
"name": "get_weather",
"description": "Get the weather in a given location",
"parameters": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": "The city and state, e.g. Chicago, IL",
},
"unit": {"type": "string", "enum": ["celsius", "fahrenheit"]},
},
"required": ["location"],
},
}
}
]
messages = [{"role": "user", "content": "What's the weather like in Chicago today?"}]
response = client.chat.completions.create(
model="google/gemini-3.5-flash",
messages=messages,
tools=tools,
tool_choice="auto"
)
print(response)
Compreensão de imagens
Os modelos do Gemini são multimodais por padrão e oferecem suporte a muitas tarefas de visão comuns.
Python
from google.auth import default
import google.auth.transport.requests
import base64
from openai import OpenAI
# TODO(developer): Update and un-comment below lines
# project_id = "PROJECT_ID"
# location = "global"
# Programmatically get an access token
credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
credentials.refresh(google.auth.transport.requests.Request())
# OpenAI Client
client = openai.OpenAI(
base_url=f"https://aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/openapi",
api_key=credentials.token,
)
# Function to encode the image
def encode_image(image_path):
with open(image_path, "rb") as image_file:
return base64.b64encode(image_file.read()).decode('utf-8')
# Getting the base64 string
# base64_image = encode_image("Path/to/image.jpeg")
response = client.chat.completions.create(
model="google/gemini-3.5-flash",
messages=[
{
"role": "user",
"content": [
{
"type": "text",
"text": "What is in this image?",
},
{
"type": "image_url",
"image_url": {
"url": f"data:image/jpeg;base64,{base64_image}"
},
},
],
}
],
)
print(response.choices[0])
Compreensão de áudio
O exemplo a seguir mostra como analisar a entrada de áudio:
Python
from google.auth import default
import google.auth.transport.requests
import base64
from openai import OpenAI
# TODO(developer): Update and un-comment below lines
# project_id = "PROJECT_ID"
# location = "global"
# Programmatically get an access token
credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
credentials.refresh(google.auth.transport.requests.Request())
# OpenAI Client
client = openai.OpenAI(
base_url=f"https://aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/openapi",
api_key=credentials.token,
)
with open("/path/to/your/audio/file.wav", "rb") as audio_file:
base64_audio = base64.b64encode(audio_file.read()).decode('utf-8')
response = client.chat.completions.create(
model="google/gemini-3.5-flash",
messages=[
{
"role": "user",
"content": [
{
"type": "text",
"text": "Transcribe this audio",
},
{
"type": "input_audio",
"input_audio": {
"data": base64_audio,
"format": "wav"
}
}
],
}
],
)
print(response.choices[0].message.content)
Resposta estruturada
Os modelos do Gemini podem gerar objetos JSON em qualquer estrutura que você definir.
Python
from google.auth import default
import google.auth.transport.requests
from pydantic import BaseModel
from openai import OpenAI
# TODO(developer): Update and un-comment below lines
# project_id = "PROJECT_ID"
# location = "global"
# Programmatically get an access token
credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
credentials.refresh(google.auth.transport.requests.Request())
# OpenAI Client
client = openai.OpenAI(
base_url=f"https://aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/openapi",
api_key=credentials.token,
)
class CalendarEvent(BaseModel):
name: str
date: str
participants: list[str]
completion = client.beta.chat.completions.parse(
model="google/gemini-3.5-flash",
messages=[
{"role": "system", "content": "Extract the event information."},
{"role": "user", "content": "John and Susan are going to an AI conference on Friday."},
],
response_format=CalendarEvent,
)
print(completion.choices[0].message.parsed)
Limitações atuais
Considere a seguinte limitação ao usar a API compatível com a OpenAI:
- Os tokens de acesso ficam ativos por 1 hora por padrão. Após a expiração, eles precisam ser atualizados. Para mais informações, consulte Atualizar suas credenciais.
A seguir
Use as bibliotecas de IA generativa do Google para chamar modelos do Gemini diretamente.
Confira mais exemplos usando a API Chat Completions com a sintaxe compatível com a OpenAI.
Consulte os modelos e parâmetros do Gemini compatíveis em Como usar bibliotecas da OpenAI com a Agent Platform.