Foundations

How models read text, how much fits in one request, and how you state the task.

9 cards on this track are live. Concept cards are free, including the ones that are not written yet. Names without a card are not links. Practice for this track is the Foundations pack. The token budget set is on that page. Context layout and prompt rewrite are not written yet.

Tokens Live Free card
The unit models read, write, and bill in.
Context Window Live Free card
The fixed text one request can hold: instructions, history, documents, and the answer.
Prompting Live Free card
How you state the task this turn: the reader, the shape, and what to leave out.
Prompt Caching Live Free card
Reusing an unchanged prefix so a repeated prompt can cost less than sending it from scratch.
Model Selection & Routing Live Free card
Picking a model, and sometimes a route between models, for the job in front of you.
Reasoning/Effort Coming soon Free card
A control some models offer for how much thinking to spend on a request.
Structured Outputs Live Free card
Asking for a reply that follows a shape, such as fields you name, so code can read it.
Hallucination Live Free card
A reply that states something the prompt and the sources do not support.
Evaluation basics Live Free card
Checking a model on examples you already know the answer to, before you trust a change.
Temperature Live Free card
A sampling control that shifts how often less likely next tokens are picked.
Sampling Coming soon Free card
How the next token is chosen from the model's probabilities.
Embeddings Coming soon Free card
Turning text into a list of numbers so similar texts land near each other.
Tokenization Coming soon Free card
The specific rules that split text into tokens. Those rules differ by model.
Multimodal inputs Coming soon Free card
Images, audio, and other non-text inputs a model can take alongside text.
Model versions Coming soon Free card
A named model release. A newer version can change behavior and tokenization.

Foundations pack Pricing