Internship← AI Product Building learning pathAI Product Building · Module 2 of 10
Choosing a model
Pick a model by quality, speed, cost and limits for your task.
📖 Key ideas
- Bigger models are usually smarter but slower and costlier.
- Smaller, fast models suit simple classification or high volume.
- Context window limits how much text you can send at once.
- Pricing is usually per million input and output tokens.
- Test several models on your own examples before choosing.
💡 Real-life example: Like choosing transport: a flight is fast and expensive, a bus is cheap and slower; pick by trip, not by habit.
TokenA small chunk of text the model reads or writes
Context windowMaximum text a model can take at once
LatencyTime until the answer arrives
ThroughputHow many requests can be handled