Splitting agent work can cut token costs sharply
Long AI work can become expensive because each later turn may reread a large amount of earlier context. In this case, the work was split between a Sonnet builder agent and a planning agent instead of being handled inside one long conversation.
The builder used about 154,000 tokens, and the planning agent used about 85,000 tokens, each in a fresh context. Doing the same work inline would likely have meant about 100 spread across 40 to 50 turns, with each turn rereading about 440,000 tokens of accumulated context.
Using a price-weighted estimate, the split-agent setup avoided about 1.5 million to 2 million weighted tokens while spending about 240,000 tokens plus review . The claimed saving for the build phase was roughly 6 to 8 times.
Key points
- Long conversations can waste tokens by rereading old context on every turn.
- Splitting work into builder and planning agents can keep each context smaller.
- The example estimates about 240,000 tokens spent versus 1.5 million to 2 million weighted tokens avoided.
- The claimed build-phase saving was about 6 to 8 times.