Q: What is gravity?
A: A force that attracts objects with mass.

Q: Explain 'quantum entanglement'.
A: Particles linked so state changes instantly.

→ generate answer for: "What is a black hole?" output: region with infinite density …

# prompt: "function to compute fibonacci recursively"
def fib(n):
    if n <= 1: return n
    return fib(n-1) + fib(n-2)

👉 also: explain this code, add types, refactor

"a serene library in space, nebula clouds, hyperrealistic, digital art, 8k —ar 16:9"

negative prompt: blurry, ugly, watermark → latent diffusion

CONTEXT (retrieved chunks):
"Einstein won Nobel prize for photoelectric effect."

QUERY: "Why did Einstein win Nobel?"
GENERATION: "For discovery of photoelectric effect."

audio input → whisper-1 
"Um actually the weather today is ..."
→ transcription + timestamps

text‑to‑speech example: “Hello world” → 🎵 emotional voice

input: image of bar chart + "summarize trends"
output: "Sales increased in Q2 by 12% ..."

Q: A ball costs $1.10. Bat costs $1 more. 
How much ball?
A: Let's think step by step... 

→ total = 1.10, bat = ball+1 → 2*ball+1=1.10 → ball=0.05

{"messages": [
  {"role": "user", "content": "tag this review"},
  {"role": "assistant", "content": "positive"}]}

LoRA rank=8 → train only 0.1% params

vector = embed("generative ai tutorials")
db.similarity_search(vector, k=3)

"golden retriever playing in ocean, cinematic"
→ 4s clip, consistent style

system: "You are a helpful botanist."
user: "Why are leaves green?"
assistant: "Chlorophyll reflects green..."

{"brand": "Nike", "items": 3, "in_stock": true}

use response_format={ "type": "json_object" }