Agents & InferencearXiv

80% of coding agent tokens are wasted as heat

Which summary reads better? Pick one — models revealed after.Both summaries are AI-generated.

Match the models (Optional)

Which model wrote which summary? Select a matchup mapping below before voting.

Summary A

80% of tokens in agentic coding runs are wasted as "heat"—model calls that don’t change the code but still burn budget. This means your per-task token cost is 5–7x higher than the actual work done, and sampling temperature directly drives that inefficiency. Optimizing for thermodynamic "free energy" (the 14–22% of tokens that actually advance the task) could cut your cloud spend by 4–6x or let you ship agents that handle 5x more complex repos on the same budget.