What You Need to Know
This objective names three model families: Haiku, Sonnet and Opus. Each is a tier, and the tiers make a real trade-off of speed and cost against depth. The Haiku class is the lowest-cost tier. Anthropic lists it as the fastest and describes it as built for high-volume, latency-sensitive tasks. The Opus class is the most capable reasoning tier, for complex, high-stakes work. The Sonnet class offers the best combination of speed and intelligence: the balanced middle and the sensible everyday default. No tier is "the good one". Each is the right choice for work that matches its strengths.
A florist has two jobs: standardising three hundred product listings into one website template, and reconciling two wholesalers' reports that partly disagree into one ordering recommendation.
The Trade-Off
The capable end takes longer per response, and model choice is among the factors that determine how quickly a usage allowance is consumed. The fast end buys its speed by not being the tool for the most demanding work. On Opus, every florist listing would take longer for no gain.
What Capability Means
Capability means depth on demanding work: untangling ambiguity and contradiction, multi-step reasoning, synthesis across sources that partly disagree, and drafting where a subtle wrong note carries real consequences. It does not mean better at everything, so the top tier is not the universal choice. For routine, well-specified work, the lighter tiers produce what the task needs, faster. The florist's listings are routine; the wholesaler recommendation is not.
Never claim a lighter tier matches a heavier one's quality on demanding work; the documentation does not. A lighter tier earns its place through fitness for the task plus review, never through equivalence.
Reading the Signals in a Task
Lighter-tier signals are high volume, a routine and repetitive shape, simple transformation such as reformatting, extracting or standardising, speed that matters, and outputs that are individually low-stakes. They are still spot-checked, because the review discipline from Domain 2 never switches off. Most-capable-tier signals are ambiguity, contradiction to untangle, judgement with consequences, complex synthesis, and work where a wrong subtle conclusion costs more than the extra time. When neither set dominates, the balanced middle is the everyday answer.
Size does not decide; the kind of work does. Standardising a long list of entries into one template is lighter-tier work however many rows it has. Reconciling two reports that partly disagree into a single recommendation is capable-tier work however short the final page. The florist's listings go to a lighter tier, sample-checked, and the one-page recommendation to Opus.
The Model Picker
You choose the model in the chat interface, not in technical configuration. Click the model name and choose another available model. You can change it at any point in a conversation, and the change applies from Claude's next response. The picker shows the models your account has access to, which varies by plan and, in workplaces, by what an organisation has enabled.
So the tier decision is not a one-off commitment. Routine ground-clearing can run on a lighter tier and the hard synthesis on the most capable, inside one piece of work. The florist can extract each wholesaler's terms on Haiku, then switch to Opus for the recommendation. Never opening the picker serves the average task and short-changes both ends. A standing default is a fact about your account, not a decision about your work.
In the exam
The guide names the three families. The objective exists because no tier suits every task. It tests reading a task's actual demands, not its prestige, a skill that outlasts any line-up. In the guide's own sample item, a high volume of short, routine drafts, where speed and cost matter more than deep reasoning, is keyed to the faster, lower-cost model. "Always use the top model" is a named wrong answer.
The prestige trap
"This project is important" does not move a routine task to the top tier. Importance belongs to projects; demands belong to tasks. The exam keys the tier that matches the task in front of you. A routine reformat inside a flagship project is still a routine reformat.
Specific model names and the current line-up change over time and belong on the quick reference.
Exam Traps
Choosing the most capable model for everything because the work is important
Prestige selection. The guide's own sample rationale calls always-using-the-top-model a waste of the cost and latency budget; routine high-volume work does not need depth, and a project's importance does not make every task inside it demanding.
Sending work that plainly needs depth to the fastest tier because it is just writing
False economy. Quality demands follow the cost of being wrong: a short output that carries judgement, contradiction or consequences is exactly where the capable tier earns its time, however quick the typing would be.
Accepting a distractor that says the lighter tier gives the same quality on demanding work
The documentation never claims tier equivalence, and no keyed answer relies on it. The lighter tier is warranted by fitness for the task plus review; when a stem needs depth, the lighter option is wrong no matter how confidently it is phrased.
Practice Scenario
A community energy co-operative employs one administrator, who has two jobs side by side: standardising six hundred member records into one address format, and drafting the co-op response to a complex regulatory consultation where three board members have given partly contradictory steers. She must keep spending on the routine job as low as she can, without risking the quality of the co-op's position. How should she assign models?
Build Exercise
Match a Week of Tasks to Tiers
What you'll learn
- How to label tasks as routine-at-volume, everyday middle, or high-stakes synthesis
- How to check which models your own account offers in the picker
- How to run a routine task on a lighter tier with a sampled quality check
- How to name what depth buys on a genuinely demanding task
- List the Claude tasks from your last working week and label each one: routine-at-volume, everyday middle, or high-stakes synthesis
Why: The three labels are the signal sets exam stems use; practising them on your own work makes them fast to spot.
You should see: A short list of real tasks, each with one of the three labels.
- For each task, note the tier its demands call for and the tier you actually used, then open the model picker and confirm which models your account offers
Why: The gap between called-for and actually-used is where the one-setting habit shows itself.
You should see: Two tier columns per task, plus the list of models your picker currently shows.
- Pick one routine task and rerun it on the lighter tier, spot-checking a sample of the output against your quality bar
Why: Fitness for the task plus review is the lighter tier warrant; the spot-check is what makes the choice professional rather than merely cheap.
You should see: Output that passes your sampled check, produced faster than your usual setting.
- Pick your most demanding recent task and write one sentence on what the capable tier bought, or would have bought, that speed could not
Why: Naming the benefit in a sentence tests whether the task genuinely needed depth or just felt important.
You should see: One sentence naming a concrete benefit: a contradiction untangled, a nuance held, a consequence avoided.
- Flag any task where you chose by prestige rather than by the task demands, and relabel it with the tier it actually needed
Why: Prestige selection is the recurring wrong answer in this objective; catching it in your own week inoculates you against it in stems.
You should see: At least one honest relabel, or a week that genuinely had none.
Sources
- Models overview — Anthropic
- Change the model, effort, and thinking settings — Anthropic
- Claude Certified Associate – Foundations Exam Guide — Anthropic