Model-model ing sawijining keluarga ngganti kapabilitas marang biaya lan latency: model sing luwih cilik luwih cepet lan murah nanging kurang kapabel; model sing luwih gedhe bisa nalar luwih apik ing masalah sing angel nanging luwih larang lan responsif luwih alon. Skill-e yaiku nyamtokake model menyang kesulitan task, ora kudu nggoleki sing paling gedhe.
Manut keluarga Claude Anthropic minangka conto (generasi paling anyar: , , ):
Haiku → smallest, fastest, cheapest
→ high-volume / simple tasks: classification, extraction, quick edits
Sonnet → balanced cost/quality
→ everyday coding and general work (the sensible default)
Opus → most capable, highest cost & latency
→ hardest multi-step reasoning, architecture, tricky debugging
Urut-urutane relatif: Haiku < Sonnet < Opus ing biaya lan latency, lan urutan sing sama ing kapabilitas mentah.
Classify 1M support tickets → Haiku (volume dominates; task is simple)
Implement a feature → Sonnet (balanced everyday coding)
Debug a subtle race condition → Opus (hard reasoning justifies the cost)
Nggunakake model sing paling gedhe kanthi default nyia-nyiake dhuwit lan nambah latency ing task sing tier sing luwih murah bisa tangani kanthi bener; nggunakake sing paling cilik kanthi default menehi asil sing ora apik ing karya sing isin banget. Milih kanthi sengaja — mulai luwih murah, naikake mung kapan kualitas njaluk — gawe sistem cepet lan murah ing skala gedhe sambil tetep entuk model abot ing masalah sing bener-bener butuh.