Token Economics
Definition
The cost structure of operating language model-based agents, measured in tokens consumed (input and output) per task. Token economics encompasses: prompt cost (system prompt + context per call), reasoning cost (chain-of-thought tokens), tool interaction cost (tool call + result tokens per loop iteration), and total task cost (all tokens across all iterations). Understanding token economics is essential for making agents commercially viable — the difference between a well-optimized and poorly-optimized agent can be 10-100x in per-task cost.
Builder Context
Track cost-per-task, not cost-per-token. A cheaper model that takes 3x more iterations costs more than an expensive model that finishes in one shot. Optimization levers: (1) reduce system prompt size (every token is paid on every call); (2) summarize conversation history instead of passing full transcripts; (3) use cheaper models for routing and simple decisions; (4) cache tool results that don't change between calls; (5) set max iteration limits to prevent runaway costs. For pricing your agent-powered product: aim for task cost < 10% of the value delivered.