CLAIM-2026-08-16-09

7% of calls went to the frontier model and accounted for 68.4% of total spend, while Nemotron 3.5 Lightning handled 93% of calls for 10.4% of spend; the judge model consumed a further 21.2% of routed spend.

Why it matters

Escalated calls are ~10x the average token weight. The requests leaving for the frontier provider are the long, context-heavy ones — which is also where the sensitive payload usually is.

Sources

• langchain2026how