Module 11 Quiz: LLM Deployment

---

Question 1

What is vLLM's PagedAttention?

1. A new attention mechanism
2. Memory management technique that dynamically allocates KV cache in pages
3. A documentation format
4. A pagination API

---

Question 2

What does INT4 quantization reduce?

1. Only speed
2. Model memory footprint (to ~12.5% of original)
3. Number of parameters
4. Training time

---

Question 3

What is the benefit of semantic caching for LLMs?

1. Faster training
2. Avoid redundant API calls by caching similar queries
3. Better model quality
4. Reduced model size

---

(Additional questions...)

---

🎉 Quiz Complete!

Next: Module 12: LLM Evaluation