Google AI API (Gemini)

pip install google-generativeai          # Google AI Studio SDK
pip install google-cloud-aiplatform      # Vertex AI SDK

import google.generativeai as genai

genai.configure(api_key="GOOGLE_API_KEY")  # or GOOGLE_API_KEY env var

model = genai.GenerativeModel(
    model_name="gemini-2.5-pro",
    system_instruction="You are a helpful assistant.",
)

# Simple generation
response = model.generate_content("Explain attention mechanisms.")
print(response.text)

# With generation config
response = model.generate_content(
    "Write a haiku about transformers.",
    generation_config=genai.GenerationConfig(
        temperature=0.9,
        max_output_tokens=100,
    ),
)

for chunk in model.generate_content("Tell me a story.", stream=True):
    print(chunk.text, end="", flush=True)

chat = model.start_chat(history=[])

response = chat.send_message("What is RAG?")
print(response.text)

response = chat.send_message("How does it compare to fine-tuning?")
print(response.text)

# Access history
for message in chat.history:
    print(f"{message.role}: {message.parts[0].text[:100]}")

import PIL.Image

model = genai.GenerativeModel("gemini-2.5-pro")

# Image from file
image = PIL.Image.open("diagram.png")
response = model.generate_content(["Explain this architecture diagram:", image])

# Image from URL
response = model.generate_content([
    "What's in this image?",
    {"mime_type": "image/jpeg", "data": base64_image_bytes},
])

# PDF analysis (Gemini handles PDFs natively)
with open("contract.pdf", "rb") as f:
    pdf_data = f.read()

response = model.generate_content([
    {"mime_type": "application/pdf", "data": pdf_data},
    "Summarise the key terms and obligations in this contract.",
])

def get_stock_price(ticker: str) -> dict:
    """Get current stock price."""
    return {"ticker": ticker, "price": 185.42, "currency": "USD"}

# Define tool
get_stock_tool = genai.protos.Tool(
    function_declarations=[
        genai.protos.FunctionDeclaration(
            name="get_stock_price",
            description="Get the current stock price for a given ticker symbol.",
            parameters=genai.protos.Schema(
                type=genai.protos.Type.OBJECT,
                properties={
                    "ticker": genai.protos.Schema(
                        type=genai.protos.Type.STRING,
                        description="Stock ticker symbol, e.g. AAPL, GOOGL",
                    )
                },
                required=["ticker"],
            ),
        )
    ]
)

model = genai.GenerativeModel("gemini-2.5-pro", tools=[get_stock_tool])
response = model.generate_content("What's Apple's current stock price?")

# Handle tool call
if response.candidates[0].content.parts[0].function_call:
    fc = response.candidates[0].content.parts[0].function_call
    result = get_stock_price(**dict(fc.args))
    
    # Send result back
    response = model.generate_content([
        "What's Apple's stock price?",
        response.candidates[0].content,
        genai.protos.Content(
            parts=[genai.protos.Part(
                function_response=genai.protos.FunctionResponse(
                    name=fc.name, response=result
                )
            )],
            role="function",
        ),
    ])

import json

model = genai.GenerativeModel(
    "gemini-2.5-flash",
    generation_config={"response_mime_type": "application/json"},
)

response = model.generate_content(
    "List the top 3 Python web frameworks with a brief description each. Return as JSON array."
)
data = json.loads(response.text)

result = genai.embed_content(
    model="models/text-embedding-004",
    content="What is the capital of France?",
    task_type="retrieval_query",  # or "retrieval_document", "semantic_similarity"
)
embedding = result["embedding"]  # list of 768 floats

import vertexai
from vertexai.generative_models import GenerativeModel

vertexai.init(project="my-gcp-project", location="us-central1")

model = GenerativeModel("gemini-2.5-pro")
response = model.generate_content("Explain quantum computing.")
print(response.text)

gcloud auth application-default login

from langchain_google_genai import ChatGoogleGenerativeAI, GoogleGenerativeAIEmbeddings

llm = ChatGoogleGenerativeAI(
    model="gemini-2.5-pro",
    google_api_key="GOOGLE_API_KEY",
    temperature=0.7,
)

embeddings = GoogleGenerativeAIEmbeddings(
    model="models/text-embedding-004",
    google_api_key="GOOGLE_API_KEY",
)

model = genai.GenerativeModel("gemini-2.5-pro")
response = model.generate_content(
    "Prove that there are infinitely many prime numbers.",
    generation_config=genai.GenerationConfig(
        thinking_config=genai.ThinkingConfig(thinking_budget=5000)
    ),
)
# response includes thinking steps + final answer