CCDV-F · Study Guide

← Study guide

Domain 5 · 16.8% of the CCDV-F exam

Model Selection and Optimization

Choose how a call is generated, how it reaches the API, which family serves it, and what the tokens cost: fundamentals, the REST client, Opus versus Sonnet versus Haiku, and caching.

Task statements

  1. 5.1 · 5.2%

    LLM Fundamentals

    Claude generates one next token at a time inside a finite context window. Sampling makes that choice vary. Thinking, effort, fast mode, and examples change how many tokens that process spends and what shape the text takes.

  2. 5.2 · 6.1%

    Technical Fundamentals

    The Claude API is HTTP. Official SDKs wrap that REST surface with types, retries, and streaming helpers. Token streaming is server-sent events on that HTTP response. A WebSocket is a different, two-way connection, and it is not how the Messages API delivers tokens.

  3. 5.3 · 2.7%

    Model Selection and Tradeoffs

    Haiku, Sonnet, and Opus are a speed, quality, and price ladder. Effort and fast mode move a call along that ladder without a new name. A new model id is a behavior change: pin the snapshot you evaluated, and rerun the eval before you serve it.

  4. 5.4 · 2.8%

    Cost and Token Management

    The bill is the usage object times the price card. Output and uncached input dominate. Prompt caching stores a shared prefix at a cache checkpoint so later calls read it at a fraction of the input price. The checkpoint only hits when that prefix, including thinking and effort, stays the same.

Build exercises