Skip to content

Browse catalog

Search catalog

[READY] Type a title, tag, or description.

NOTES // concept

Worker-tier routing

LAST_MODIFIED
2026.07.29
CATEGORY
concept

Worker-tier routing sends planning to a strong model and execution to a cheaper tier. Quality lives in the plan; the executor runs the spec.

Tier$/M in/out (approx.)Role
Fable 5$10 / $50Rare hardest-step escalation (refusals)
Opus 5$5 / $25Long-horizon orchestration
Sonnet 5$2 / $10 intro → $3 / $15Default worker
Haiku 4.5$1 / $5Parallel sub-agents
DeepSeek V4 Flash$0.14 / $0.28Cheap executor (complete spec required)

DeepSeek V4 Flash scores 79.0 on SWE-bench Verified against Opus 5’s 96.0, and 60/100 on a real agentic build - strong at executing a complete plan, weak when inventing architecture. One Flash output token costs ~1/89th of an Opus 5 token. Practitioners calibrate it as Sonnet-class for clear specs, not Opus-class for vague ones.

Conditions that make the split work

  1. The plan is genuinely complete - every ambiguity becomes an invention.
  2. The prompt prefix stays stable - cache hits are where the economics live.
  3. Still verify before done; a cheap worker is not an excuse to skip checks.
  4. Effort vs fast mode are separate dials on Claude tiers - do not confuse them with model choice.

Evidence: DeepSeek V4 Flash is a worker tier, not an Opus killer. Stack map: Why most agents default to the wrong Claude tier.

model-selectioncostagentsrouting

Related_Notes