Gen-AI Developer Classroom notes 28/Jul/2026

Models

• We have two types of models
  • Open source Models
    • OpenAI
    • Qwen
    • GLM
    • Google Gemma
  • Vendor Models:
    • Popular:
      • Claude
      • Open AI
      • Gemini
      • Amazon
      • Groq
• To access open source models we have following options
  • Run on your hardware
  • Hosted:
    • Cloud
      • GCP
      • AWS
      • Azure

• To access vendor specific models

  • Direct Vendor APIs
  • Cloud Hosted:
    • AWS
    • Azure
    • GCP

• Models by Modal

  • Text
  • Code
  • Image
  • Audio
  • Video

Accessing models

• We need to access models programatically

Exercise: 1. Accessing models via code

• To interact with models we need the following
  • Prompt
  • credentials (API key)
  • SDK
• Navigate to google ai studio and login with your google account and create an API Key

• Prompt: What is capital of France

• Lets try from language python

• Note: Watch classroom video for curl and code based access of models from gemini.

Exercise: 2.

• I have a selfie convert that into a passport size photograph for some job application

• For this We need a model which understands images (Vision Model)

Exercise 3

• For some reason i want to compose music, a special song for Spain for winning FIFA

• Tech stack

  • Speech to TExt
  • Text To speech
  • Vocal library
  • Music generation models
    • metagen
    • lyria 3

Understanding model

• A Large Language model generates token’s repeatedly till the
  • limit is reached
  • token is found
• If the model is just predicting next token how is it intelligent?