Generative AI Google Brain

gemini-use-cases-education-use-cases-for-education

3 min read

Using Gemini in Education

Source notebook

Repo path: gemini/use-cases/education/use_cases_for_education.ipynb · Open on GitHub · intro

Demonstrates Gemini education prompts across text, math, images, multiple images, and video.

Summary

This notebook teaches how to use the Google Gen AI SDK with the Gemini API in Vertex AI for education-oriented tasks. It initializes a Vertex AI-backed GenAI client, defines a small generation helper, and runs prompts for reasoning, summarization, translation, correction, math, image understanding, multi-image comparison, and video understanding. The workflow shows text-only and multimodal calls using URL and Cloud Storage media parts.

Key code patterns

Initialize Vertex AI GenAI client

from google import genai
 
PROJECT_ID = os.environ.get("GOOGLE_CLOUD_PROJECT")
LOCATION = os.environ.get("GOOGLE_CLOUD_REGION", "us-central1")
client = genai.Client(vertexai=True, project=PROJECT_ID, location=LOCATION)

Uses the Google Gen AI SDK against Vertex AI with project and region settings.

Reusable content generation helper

def generate_content(model_id: str, contents: list | str) -> str:
    return client.models.generate_content(
        model=model_id,
        contents=contents,
        config=GenerateContentConfig(max_output_tokens=8192),
    ).text

Centralizes Gemini calls and sets a maximum output token limit for responses.

Image input from URI

image_abacus = Part.from_uri(
    file_uri="https://images.unsplash.com/photo-1548690596-f1722c190938?...",
    mime_type="image/jpeg",
)
contents = ["Describe this image in a short sentence:", image_abacus]

Shows how to pass remote image content with text prompts for visual reasoning.

Video input from Cloud Storage

video = Part.from_uri(
    file_uri="gs://cloud-samples-data/video/animals.mp4",
    mime_type="video/mp4",
)
contents = [prompt, video]

Shows Gemini video understanding with timestamped questions over a GCS video file.

Models & APIs used

  • Models: gemini-3.5-flash
  • APIs / services: Vertex AI, Gemini API in Vertex AI, Cloud Storage
  • SDKs / libraries: google-genai

When to use this

Use this pattern to prototype education workflows that need Gemini text, image, and video reasoning through Vertex AI.

Gotchas & caveats

  • This tutorial uses billable Vertex AI components.
  • Colab requires explicit Google Cloud authentication; Vertex AI Workbench does not.
  • PROJECT_ID must be set directly or available as GOOGLE_CLOUD_PROJECT.
  • LOCATION defaults to us-central1 when GOOGLE_CLOUD_REGION is unset.
  • The helper function signature shown does not accept temperature, but one later call passes temperature=1.0.
  • The notebook notes that large language models can hallucinate math results and recommends step-by-step prompts or calculator libraries for advanced math.
  • Video files are sampled at 1 frame per second, with video and audio samples analyzed with timestamps.

Best practices

  • Use a low default temperature for more consistent responses.
  • Use few-shot examples to guide response structure and formatting for text correction.
  • Ask for step-by-step reasoning to reduce hallucinations in math tasks.
  • Use structured output requests such as JSON lists or tables when asking detailed questions.
  • Ask video questions using the video only and request timestamps plus source type such as image, text, or speech.

Graph View

Backlinks