Quick Reference: Domain 3 — Product and Model Selection

Read the task's shape before writing any prompt. Four questions decide it: one-off or recurring? Standalone, or needing the same standing context? Consumed once, or revised and reused? Answerable from supplied material? If not, one fact from outside, or a report built from many outside sources?

On the exam: find the dominant signal first, then match it. Ordinary chat is a legitimate right answer, and the most elaborate feature is not the safest choice. Re-pasting the same context chat after chat is doing a Project's job by hand; Research adds nothing to a question the attached material already answers. Instructions and knowledge curation belong to Domain 5; format choice for a single output belongs to Domain 2.

The exam guide names three families, and the tested skill is reading which end of the trade-off a task needs:

• Haiku class: the fastest, lowest-cost tier, which Anthropic describes as built for high-volume, latency-sensitive tasks. Best suited to: high volume, routine shape, simple transformation, spot-checked outputs.
• Sonnet class: the best combination of speed and intelligence; the balanced choice for everyday tasks when neither signal set dominates.
• Opus class: the most capable reasoning tier, for ambiguity, contradiction, complex synthesis and judgement with consequences.

Never claim tier equivalence: the documentation nowhere says a lighter tier matches a heavier one on demanding work. What justifies the lighter tier is fitness for the task plus review. A reviewer can catch errors that are on the page. A reviewer cannot add reasoning that the draft does not contain, so review makes a lighter tier safe only when the task is routine.

The picker: click the model name in the chat interface; the change applies from the next response, and the picker shows what your account offers. Mid-conversation switching means the decision is never a one-off commitment.

Current line-up (checked 13 November 2026): the models overview lists Claude Fable 5.2 and Claude Opus 5.5 at the capable end, Claude Sonnet 5.5 as the balanced middle and Claude Haiku 5.5 as the fastest. For a user who is unsure which model to use, it recommends starting with Claude Opus 5.5 for most workloads. The guide's objective still names the Haiku-Sonnet-Opus triad, so answer exam questions in tiers, not version numbers.

Every model choice spends three currencies: cost, speed and quality; name the one the task needs and pay only that.

• Cost means the plan's usage allowance, not a per-use invoice. Named consumption factors: model choice, effort level, current conversation length, file attachment size.
• Volume multiplies cost and speed. Run once, tier differences are seconds; run hundreds of times, they decide whether the job finishes and the allowance lasts. Bulk work goes to the fast tier with a sampled spot-check.
• Stakes multiply quality. One consequential output needs more than hundreds of routine ones do; "it is only one letter" is true and irrelevant. High-stakes output gets the capable tier plus qualified human review, whatever model drafted it.
• Effort is a second setting in the model menu. It controls how much thinking Claude applies to a response. Low and Medium effort work well for routine tasks and make the allowance last longer. Current guidance (checked 29 October 2026): before you move to an older or smaller model only to stay inside your limits, try a lower effort level. When a question asks which tier to use, answer with the tier that the task needs, not with an effort setting. For high-volume routine work, that is the faster, lower-cost tier.
• Split mixed workloads by demands, not habit: bulk fast with sampled review, the hard core capable with proper review.

Three recurring wrong answers: maximum quality everywhere (waste grows with volume), minimum cost everywhere (shallow output exactly where errors are expensive), and the escalation loop (re-running a vague brief on ever-heavier models; a vague brief fails at every tier, and fixing it belongs to Domain 1).

A conversation holds a finite working space, and very long conversations approach the platform's length and usage limits. The official options there are practical: start a new conversation, or use a Project for larger amounts of information. Anthropic's Academy adds that the quality of Claude's answers can get worse in a very long conversation. The objective tests planning, not repair of damage.

• Restart when the work is done, when the chat has accumulated abandoned approaches and superseded drafts, or when limits approach.
• Summarise before leaving: decisions, current state, open questions, then verify the summary against source records; it contains every uncorrected error in the chat. The verified summary opens the fresh conversation.
• Persist what the next conversation needs: paste it in, re-attach the file, or store it in Project knowledge. A standing document that keeps being revised and shared belongs in an artifact.

Check the question's time frame: organising work for what comes next is 3.4; diagnosing a chat already producing poor output is Domain 7. "A restart would lose all our context" is the marathon chat's argument, and it is false: the context worth keeping is exactly what the handover moves to the new conversation.

• Rollout (dated): cross-chat memory arrived on claude.ai on 11 September 2025 for Team and Enterprise plans. It was extended to Pro and Max from 23 October 2025. As of 14 November 2026, it is on by default for Free, Pro, Max and Ultra plans. As of that date, on Team and Enterprise plans an owner decides whether memory is available. In organisations created before 31 October 2026, memory stays off for each member until that member enables it. In most newer ones, memory starts on.
• What memory holds: a set of individual topics covering your role, projects, preferences and ongoing work context. Context about you, not transcripts, drafts or attached documents. Each Project has its own memory, used alongside your memory from chats outside Projects but never another Project's.
• Chat search: on paid plans you can ask Claude to search past chats and reference relevant details. Prompted retrieval, not automatic carry-over.
• The boundary the exam tests: memory saves re-introducing yourself; it does not reliably transport a seating plan, an approved outline or a data file into a new chat. Two stale claims fail in opposite directions: that Claude cannot remember anything across chats, and that memory can be trusted to carry a working document. The dependable pattern for a specific working output is explicit persistence.