google-ai-studio

# Python
pip install google-generativeai

# Node.js
npm install @google/generative-ai

export GOOGLE_API_KEY=AIza...

import google.generativeai as genai

genai.configure(api_key="AIza...")  # or reads GOOGLE_API_KEY

model = genai.GenerativeModel("gemini-2.0-flash")
response = model.generate_content("Explain neural networks in one paragraph.")
print(response.text)

import google.generativeai as genai

genai.configure(api_key="AIza...")

model = genai.GenerativeModel(
    model_name="gemini-2.0-flash",
    system_instruction="You are a Python expert. Always show working code examples.",
)

chat = model.start_chat()
response = chat.send_message("How do I read a CSV with pandas?")
print(response.text)

response = chat.send_message("Now show me how to filter rows where age > 30.")
print(response.text)

import google.generativeai as genai
import PIL.Image

genai.configure(api_key="AIza...")

model = genai.GenerativeModel("gemini-2.0-flash")

# From local file
image = PIL.Image.open("screenshot.png")
response = model.generate_content(["What's in this image? List all visible text.", image])
print(response.text)

# From URL (inline data)
import httpx
import base64

img_data = httpx.get("https://example.com/chart.png").content
image_part = {"mime_type": "image/png", "data": base64.b64encode(img_data).decode()}
response = model.generate_content(["Analyze this chart:", image_part])
print(response.text)

import google.generativeai as genai
import pathlib

genai.configure(api_key="AIza...")

model = genai.GenerativeModel("gemini-1.5-pro")

# Upload a PDF file
pdf_file = genai.upload_file(
    path="report.pdf",
    mime_type="application/pdf",
    display_name="Annual Report 2024",
)

response = model.generate_content([
    "Summarize the key financial metrics from this report.",
    pdf_file,
])
print(response.text)

# Inline PDF (smaller files)
pdf_bytes = pathlib.Path("document.pdf").read_bytes()
import base64
pdf_part = {"mime_type": "application/pdf", "data": base64.b64encode(pdf_bytes).decode()}
response = model.generate_content(["Extract all dates and deadlines:", pdf_part])
print(response.text)

import google.generativeai as genai

genai.configure(api_key="AIza...")

model = genai.GenerativeModel("gemini-2.0-flash")

for chunk in model.generate_content("Write a short story about AI.", stream=True):
    print(chunk.text, end="", flush=True)
print()

import google.generativeai as genai
import json

genai.configure(api_key="AIza...")

model = genai.GenerativeModel(
    model_name="gemini-2.0-flash",
    generation_config={
        "response_mime_type": "application/json",
        "response_schema": {
            "type": "object",
            "properties": {
                "companies": {
                    "type": "array",
                    "items": {
                        "type": "object",
                        "properties": {
                            "name": {"type": "string"},
                            "founded": {"type": "integer"},
                            "country": {"type": "string"},
                        },
                        "required": ["name", "founded", "country"],
                    },
                }
            },
        },
    },
)

response = model.generate_content(
    "List 3 major AI companies with their founding year and country."
)
data = json.loads(response.text)
print(data)

import google.generativeai as genai
from google.generativeai import types

genai.configure(api_key="AIza...")

model = genai.GenerativeModel("gemini-2.0-flash")

# Enable Google Search grounding
response = model.generate_content(
    "What are the latest AI research papers published this week?",
    tools=[types.Tool(google_search=types.GoogleSearch())],
)

print(response.text)

# Check grounding metadata
if response.candidates[0].grounding_metadata:
    for source in response.candidates[0].grounding_metadata.search_entry_point or []:
        print(f"Source: {source}")

import google.generativeai as genai

genai.configure(api_key="AIza...")

def get_product_info(product_id: str) -> dict:
    """Simulated product lookup."""
    return {"id": product_id, "name": "Widget Pro", "price": 49.99, "in_stock": True}

model = genai.GenerativeModel(
    model_name="gemini-2.0-flash",
    tools=[get_product_info],  # Pass Python function directly!
)

chat = model.start_chat(enable_automatic_function_calling=True)
response = chat.send_message("What's the price and availability of product P123?")
print(response.text)
# Gemini automatically calls get_product_info("P123") and incorporates the result

import google.generativeai as genai
import pathlib

genai.configure(api_key="AIza...")

model = genai.GenerativeModel("gemini-1.5-pro")  # 2M token context

# Read entire codebase into context
files = list(pathlib.Path("./src").rglob("*.py"))
code_content = "\n\n".join([
    f"# File: {f}\n{f.read_text()}" for f in files
])

response = model.generate_content([
    "Analyze this codebase and identify security vulnerabilities:",
    code_content,
])
print(response.text)

import google.generativeai as genai

genai.configure(api_key="AIza...")

# Single embedding
result = genai.embed_content(
    model="text-embedding-004",
    content="Machine learning transforms industries.",
    task_type="retrieval_document",
)
print(f"Embedding dim: {len(result['embedding'])}")  # 768

# Batch embeddings
texts = ["Hello world", "Machine learning", "AI systems"]
result = genai.embed_content(
    model="text-embedding-004",
    content=texts,
    task_type="retrieval_document",
)
embeddings = result["embedding"]  # List of 768-dim vectors