Why AI Model Names Change So Often (Claude 4.5, GPT-5, Gemini 2.5)

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-builders-models-model-versions-r8a8-teen

1. A developer pins their app to use 'gpt-4' in their API code. What will happen when OpenAI releases GPT-5?

   1. The app will automatically start using GPT-5
   2. The app will crash because the version is invalid
   3. The app will continue using GPT-4 until the developer changes the version string
   4. The app will ask the user which model to use

2. What does it mean when an AI lab "deprecates" a model like gpt-3.5-turbo?

   1. The model becomes free to use
   2. The model gets better with age like fine wine
   3. The model is removed from the API and code calling it will eventually break
   4. The model is moved to a premium tier with no changes

3. A student wrote code last year using Claude 3.5 Sonnet that worked perfectly. Now they're testing the same code with Claude 4.5, and some prompts produce different results. Why?

   1. Claude 4.5 has a bug that makes it ignore prompts
   2. The model is a different version with different behaviors and training
   3. Claude 4.5 requires a paid subscription to work correctly
   4. The student accidentally changed their prompt text

4. A chat app like ChatGPT or Claude AI typically defaults to which model version?

   1. You cannot change the default model
   2. The latest version available
   3. A random version each time you open it
   4. The oldest available version

5. A developer wants their production app to always behave the same way for users. What should they do?

   1. Avoid using APIs entirely
   2. Pin a specific model version in their API code
   3. Use the latest model always
   4. Change models every week

6. Why might upgrading from GPT-4 to GPT-5 save a developer money?

   1. GPT-5 is described as cheaper than GPT-4 in the lesson
   2. GPT-5 uses less computational power per request
   3. OpenAI gives free credits to all users
   4. GPT-5 is open source

7. What should you do BEFORE judging whether an AI model is good or bad for your use case?

   1. Assume it will perform the same as the previous version
   2. Delete your old prompts
   3. Re-test your prompts with the new version
   4. Write a one-star review online

8. If you write a tutorial using Claude 4.5 today, what might happen to someone reading it six months later?

   1. They must pay to access the tutorial
   2. They might get different results if the model has updated
   3. The tutorial will stop working entirely
   4. Nothing will change, AI models never update

9. A company builds an app that relies on AI to summarize customer emails. They upgrade their model without re-testing their prompts. What could go wrong?

   1. Nothing will change because AI is perfect
   2. Customers will get charged more money automatically
   3. The app will stop working completely
   4. The summaries might become longer, shorter, or different in quality

10. Why does knowing the specific model version matter for cost planning?

    1. Different versions have different prices and rate limits
    2. Older models are always more expensive
    3. All models cost the same
    4. Price doesn't change between versions

11. What does it mean that a model has "different strengths" between versions?

    1. The model does pushups
    2. All versions are exactly the same
    3. Strength refers to server capacity
    4. One version might be better at coding while another is better at creative writing

12. You notice your favorite AI chat app now behaves differently than last month. What's the most likely reason?

    1. The app is on strike
    2. Your internet connection changed
    3. The app is broken
    4. The model version was updated to a newer release

13. What should a developer do when their pinned model version is announced as deprecated?

    1. Switch to a different AI company entirely
    2. Ignore it and keep using the old version
    3. Plan to upgrade to a supported version before the deprecation date
    4. Delete their entire application

14. What does the lesson suggest you do in your favorite chat app's settings?

    1. Disable all features
    2. Check which model version it currently uses
    3. Change the app's color theme
    4. Delete your account

15. A student says: 'I don't need to worry about model versions because AI always improves automatically.' What's wrong with this thinking?

    1. Nothing - the student is correct
    2. Model updates can change behavior in unexpected ways, requiring re-testing
    3. AI never improves
    4. Only adults need to worry about versions