--- title: Token Spend Out of Control? The Case for Smarter Routing description: LLM agents burn tokens fast by running in loops and resending full context on every step, often using expensive frontier models for all tasks. The solution is... canonical: https://daily.dev/posts/token-spend-out-of-control-the-case-for-smarter-routing-jxkbskjha twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Token Spend Out of Control? The Case for Smarter Routing | daily.dev og:description: LLM agents burn tokens fast by running in loops and resending full context on every step, often using expensive frontier models for all tasks. The solution is... og:url: https://daily.dev/posts/token-spend-out-of-control-the-case-for-smarter-routing-jxkbskjha og:image: https://api.daily.dev/og/posts/jxKbSKJHA.png og:image:alt: Token Spend Out of Control? The Case for Smarter Routing og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Token Spend Out of Control? The Case for Smarter Routing **[ByteByteGo](https://daily.dev/sources/bytebytego)** · 15 min read · 3 upvotes · 0 comments ## Summary LLM agents burn tokens fast by running in loops and resending full context on every step, often using expensive frontier models for all tasks. The solution is intelligent routing: directing each request to the cheapest model capable of handling it. Using Kilo's open-source coding agent as a case study, the post explains how routing works (entry point + decision layer), two routing strategies (signal-based vs. difficulty-prediction), and real production numbers showing ~33% cost reduction with 80-90% of requests not needing frontier models. Key lessons: set a fixed budget, log tokens per task type to find where spend actually lives, and route on the strongest signal you already have (e.g., task mode) before resorting to ML-based difficulty classifiers. Caching helps but doesn't solve volume problems alone — routing addresses a different dimension of cost. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [A better way to manage LLM spending](https://daily.dev/posts/a-better-way-to-manage-llm-spending-2th8qnxp7) · InfoWorld · 0 upvotes · 0 comments - [Model Routing Is Simple. Until It Isn’t.](https://daily.dev/posts/model-routing-is-simple-until-it-isn-t--hk5ukjdm8) · Hugging Face · 0 upvotes · 0 comments - [How LLM Routing Actually Works in Production](https://daily.dev/posts/how-llm-routing-actually-works-in-production-rsf5hw9mf) · Daily Dose of Data Science \| Avi Chawla \| Substack · 2 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#finops](https://daily.dev/tags/finops) [View this post on daily.dev](https://daily.dev/posts/token-spend-out-of-control-the-case-for-smarter-routing-jxkbskjha) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Token Spend Out of Control? The Case for Smarter Routing","url":"https://daily.dev/posts/token-spend-out-of-control-the-case-for-smarter-routing-jxkbskjha","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/token-spend-out-of-control-the-case-for-smarter-routing-jxkbskjha"},"datePublished":"2026-06-08T15:08:17.547Z","dateModified":"2026-06-08T15:08:40.409Z","description":"LLM agents burn tokens fast by running in loops and resending full context on every step, often using expensive frontier models for all tasks. The solution is...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/37e4cf7a499ce6f451aea703376dfa2a?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/37e4cf7a499ce6f451aea703376dfa2a?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"ByteByteGo","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"ByteByteGo","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/35be29234ee14d01a9cd049c52e12753","url":"https://daily.dev/sources/bytebytego"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/token-spend-out-of-control-the-case-for-smarter-routing-jxkbskjha","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":3},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,finops","timeRequired":"PT15M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"ByteByteGo","item":"https://daily.dev/sources/bytebytego"},{"@type":"ListItem","position":3,"name":"Token Spend Out of Control? The Case for Smarter Routing"}]} ```