OpenAI GPT-4.5 and GPT-5 Quiz

Answers and explanations appear as you take the quiz.

1. What system structure did OpenAI describe for GPT-5 at its August 2025 launch?
   • A fast model, a deeper reasoning model, and a real-time router
   • A single non-reasoning model with no routing
   • A collection of image-only generators
   • A recurrent speech recognizer
2. What was GPT-4.5 known for emphasizing?
   • Heavy step-by-step chain-of-thought reasoning
   • Broader world knowledge, fewer hallucinations, and a warmer conversational style
   • Image generation only
   • Running entirely offline on phones
3. Why does OpenAI offer mini and nano variants of its models?
   • To make the models slower on purpose
   • Because the full models cannot run in the API
   • To provide cheaper and faster options for lighter workloads
   • To remove safety features
4. What did OpenAI say distinguished GPT-4.5’s response style from reasoning models such as o1?
   • GPT-4.5 was a voice-only model
   • GPT-4.5 required an explicit multi-step reasoning trace for every answer
   • GPT-4.5 was a decision-tree system
   • GPT-4.5 did not think before responding and emphasized general-purpose knowledge and conversation
5. Which improvement was a stated goal for GPT-5 compared to earlier models?
   • Reduced hallucinations and stronger coding and agentic tool use
   • Higher hallucination rates
   • No ability to use tools
   • Smaller context windows
6. What did OpenAI’s June 2026 release note say about GPT-4.5 in ChatGPT?
   • It became the only free ChatGPT model
   • It was retired from ChatGPT on June 27, 2026
   • It was replaced by GPT-4.5 mini
   • It was retired from the API at the same time
7. At GPT-5’s launch, how did the API offering differ from GPT-5 in ChatGPT?
   • The API offered only an image model
   • ChatGPT exposed no model routing
   • The API gave direct access to a reasoning model, while ChatGPT used a routed system
   • Both products used the exact same single endpoint and interface
8. Which developer controls did OpenAI introduce with the GPT-5 API?
   • A camera calibration parameter
   • A spectrogram frequency scale
   • A robot joint-torque limit
   • Reasoning-effort and verbosity settings