Why a two word follow up in a long debugging session does not get routed to a cheap model.
Here is a problem every per prompt router has, and most of them lose to.You are twenty messages into debugging a race condition. You have pasted stack traces, the model has written three patches, and you type:
now do the rest
Scored on its own, that prompt is four trivial words. Scored honestly, it inherits everything above it. A router that only looks at the current message sends it to the cheapest model on the list, and you get a useless answer at the worst possible moment.Routor reads the conversation, not just the message.
The history floor
Before choosing a model, Routor scans the recent conversation and may set a floor: a tier the request cannot fall below, no matter how simple this particular message looks.The floor is applied as max(scored tier, floor). It can only raise a request, never lower one.
What raises it
Signal
Where
Floor
The assistant wrote a code block
assistant messages
LIGHT
You pasted an error or stack trace
user messages
STANDARD
Two or more reasoning keywords in one message
user messages
STANDARD
Eight or more messages, with any signal above already present
whole thread
STANDARD
Error detection is not keyword spotting on the word “error”. It matches real failure shapes across languages: tracebacks, panics, segfaults, borrow checker complaints, npm ERR!, ModuleNotFoundError, NullPointerException, cannot read propert..., index out of bounds.
What it reads
The last 6 conversational messages, capped at 4,000 characters each, system and tool messages excluded. The current prompt is deliberately left out, since it has already been scored on its own merits. Threads shorter than two turns are skipped entirely.
History alone never forces COMPLEX. The floor is capped at STANDARD, so a long conversation can keep you in capable territory but cannot by itself escalate you to the most expensive tier. Reaching COMPLEX requires the current prompt to earn it.
Saying “thanks” does not cost you money
The obvious failure of a floor is that it makes everything expensive forever. Type “hi” in a coding session and you would pay STANDARD rates to be told hello.So trivial messages are exempt. Anything at or under 60 characters that is recognisably an acknowledgement skips the floor and routes on its own merits:
hi hey ok sure thanks thx got it sounds goodcool nice perfect alright yep nope done bye 👍 🙏 🎉
A greeting in the middle of a long debugging session still goes to a cheap model. The floor comes back the moment you say something real.
Category carry forward
The floor decides how capable the model must be. A second, quieter mechanism decides what kind of model it should be.When history shows a coding session, Routor remembers the task category and carries it forward through thin follow ups. A vague “now the tests too” stays in the coding category instead of drifting into general chat, because the session already established what you are doing. An error trace marks the session as debugging specifically.
The reasoning string names each signal and the turn it came from, so you can tell whether it was the stack trace you pasted or the code block the model wrote that held the floor up. When no floor applies, the header is absent.See Router Metadata for the full header list.
Why this matters more for agents
Agent loops are where this pays for itself. An agent sends dozens of short, context free instructions: “continue”, “fix it”, “now the tests”. Every one of those, judged alone, looks trivial. Judged in context, most of them are the hard part of a coding task.Without conversation memory, a per prompt router quietly downgrades an agent at exactly the wrong moments, and the loop produces worse work while appearing to save money.
{"docsConfig":{"theme":"mint","$schema":"https://mintlify.com/docs.json","name":"Routor","colors":{"primary":"#16A34A","light":"#4ADE80","dark":"#15803D"},"logo":{"light":"https://mintcdn.com/routor/hv7GNqBT8JfNtoVg/logo/logo-light.svg?fit=max\u0026auto=format\u0026n=hv7GNqBT8JfNtoVg\u0026q=85\u0026s=54f11b9d71088b9ca9f3b6214afed989","dark":"https://mintcdn.com/routor/hv7GNqBT8JfNtoVg/logo/logo-dark.svg?fit=max\u0026auto=format\u0026n=hv7GNqBT8JfNtoVg\u0026q=85\u0026s=a3821333872affc217b83891fcfaa8ca"},"favicon":"https://mintcdn.com/routor/hv7GNqBT8JfNtoVg/logo/logo.png?fit=max\u0026auto=format\u0026n=hv7GNqBT8JfNtoVg\u0026q=85\u0026s=4360f319a0798dedc5a038ec5e5ab407","navbar":{"links":[{"href":"https://routor.io","label":"Website"},{"href":"https://dashboard.routor.io","label":"Dashboard"}],"primary":{"type":"button","label":"Get Started Free","href":"https://dashboard.routor.io/signup"}},"footer":{"socials":{"twitter":"https://x.com/RoutorAI","github":"https://github.com/routor-ai"}},"styling":{"latex":true},"navigation":{"pages":[]}},"deployment":{"deploymentMetadataValue":{"subdomain":"routor","hasAudioNarration":false,"actualSubdomain":"routor","customDomains":["docs.routor.io"],"gitSource":{"type":"github","owner":"buildorz","repo":"smart-router","deployBranch":"main","contentDirectory":"docs","isPrivate":true,"mountPath":"","label":""},"entitlements":{"AI_CHAT":{"status":"DISABLED"}},"buildId":"6abe8053d30dc67faef74b3f","clientVersion":"0.0.4270","searchFilterCounts":[{"language":"default","tag":"Docs","count":37},{"language":"default","tag":"Guides","count":9},{"language":"default","tag":"Getting Started","count":7},{"language":"default","tag":"Playground","count":6},{"language":"default","tag":"Core Concepts","count":5},{"language":"default","tag":"Dashboard","count":5},{"language":"default","tag":"Support","count":4},{"language":"default","tag":"Ecosystem","count":3},{"language":"default","tag":"API Reference","count":2},{"language":"default","tag":"Endpoints","count":2}]},"assistantConfig":{"enableAskAiButton":false,"enableStarterQuestions":false,"enableContextualStarterQuestions":false,"enableAskAiSearchBar":true},"basePath":"","hasChatPermissions":false,"hasWebMcpPermissions":false,"toggles":[{"name":"kill-isr","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"custom-rbac-roles","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"agent-mintlify-slack","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"agent-integrations-settings","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"rbac-expansion","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"automations-pr-grouping","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"custom-prefix","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"analytics-v7","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"user-flows-analytics","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"no-publish-flow","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workos-onboarding","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"snapshot-serving-live","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"stripTemperatureMiddleware","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"support-platform-pylon-zendesk","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"snapshot-serving","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"instant-rollback","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"granular-notifications-server","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"granular-notifications","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"automations-fable","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"fixed-ai-action-pricing","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"site-roles","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"automation-resolve-merge-conflict","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"scoped-publishing","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"mintql","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"page-level-analytics","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"editor-version-history-v2","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"billing-page-v2","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"full-page-search","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workspace-projects","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"assistant-v5","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"fork-pr-previews","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"revalidation-jwt","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workos-sso-scim","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"hobby-abuse-scan","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"dashboard-inbox","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workos-signup","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"turbopuffer-dual-write","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workos-enabled","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"approve-review-flow","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"editor-metadata","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"sso-setup-deadline-banner","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"turbopuffer-serve","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"live-agents","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"starter-pricing","enabled":true,"variant":{"name":"control","enabled":true,"payload":{"type":"string","value":"control"},"feature_enabled":true},"impressionData":false},{"name":"llms-txt-v2","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"wildcard-email-auth","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false}],"googleFontsInlined":false},"customCss":[]}
{"page":{"pagePath":"conversation-memory","slug":"/conversation-memory","pageMetadata":{"title":"Conversation Memory","description":"Why a two word follow up in a long debugging session does not get routed to a cheap model.","icon":"comments","href":"/conversation-memory"},"description":{"compiledSource":"\"use strict\";\nconst {jsx: _jsx} = arguments[0];\nconst {useMDXComponents: _provideComponents} = arguments[0];\nfunction _createMdxContent(props) {\n const _components = {\n p: \"p\",\n ..._provideComponents(),\n ...props.components\n };\n return _jsx(_components.p, {\n children: \"Why a two word follow up in a long debugging session does not get routed to a cheap model.\"\n });\n}\nfunction MDXContent(props = {}) {\n const {wrapper: MDXLayout} = {\n ..._provideComponents(),\n ...props.components\n };\n return MDXLayout ? _jsx(MDXLayout, {\n ...props,\n children: _jsx(_createMdxContent, {\n ...props\n })\n }) : _createMdxContent(props);\n}\nreturn {\n default: MDXContent\n};\n","frontmatter":{},"scope":{}},"pageType":"default","mdx":{"mdxSource":{"compiledSource":"\"use strict\";const{Fragment:_Fragment,jsx:_jsx,jsxs:_jsxs}=arguments[0];const{useMDXComponents:_provideComponents}=arguments[0];function _createMdxContent(props){const _components={a:`a`,code:`code`,hr:`hr`,p:`p`,pre:`pre`,span:`span`,strong:`strong`,tbody:`tbody`,td:`td`,th:`th`,thead:`thead`,tr:`tr`,..._provideComponents(),...props.components},{CodeBlock,Heading,Note,Table}=_components;if(!CodeBlock)_missingMdxReference(`CodeBlock`,true);if(!Heading)_missingMdxReference(`Heading`,true);if(!Note)_missingMdxReference(`Note`,true);if(!Table)_missingMdxReference(`Table`,true);return _jsxs(_Fragment,{children:[_jsx(_components.p,{children:`Here is a problem every per prompt router has, and most of them lose to.`}),`\n`,_jsx(_components.p,{children:`You are twenty messages into debugging a race condition. You have pasted stack traces, the model has written three patches, and you type:`}),`\n`,_jsx(CodeBlock,{filename:``,numberOfLines:`2`,language:`text`,children:_jsx(_components.pre,{className:`shiki shiki-themes github-light-default dark-plus`,style:{backgroundColor:`#ffffff`,\"--shiki-dark-bg\":`#0B0C0E`,color:`#1f2328`,\"--shiki-dark\":`#D4D4D4`},language:`text`,children:_jsxs(_components.code,{language:`text`,numberOfLines:`2`,children:[_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:`now do the rest`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{})})]})})}),`\n`,_jsx(_components.p,{children:`Scored on its own, that prompt is four trivial words. Scored honestly, it inherits everything above it. A router that only looks at the current message sends it to the cheapest model on the list, and you get a useless answer at the worst possible moment.`}),`\n`,_jsx(_components.p,{children:`Routor reads the conversation, not just the message.`}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`the-history-floor`,children:`The history floor`}),`\n`,_jsxs(_components.p,{children:[`Before choosing a model, Routor scans the recent conversation and may set a `,_jsx(_components.strong,{children:`floor`}),`: a tier the request cannot fall below, no matter how simple this particular message looks.`]}),`\n`,_jsxs(_components.p,{children:[`The floor is applied as `,_jsx(_components.code,{children:`max(scored tier, floor)`}),`. It can only raise a request, never lower one.`]}),`\n`,_jsx(Heading,{level:`3`,id:`what-raises-it`,children:`What raises it`}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Signal`}),_jsx(_components.th,{children:`Where`}),_jsx(_components.th,{children:`Floor`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:`The assistant wrote a code block`}),_jsx(_components.td,{children:`assistant messages`}),_jsx(_components.td,{children:_jsx(_components.code,{children:`LIGHT`})})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:`You pasted an error or stack trace`}),_jsx(_components.td,{children:`user messages`}),_jsx(_components.td,{children:_jsx(_components.code,{children:`STANDARD`})})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:`Two or more reasoning keywords in one message`}),_jsx(_components.td,{children:`user messages`}),_jsx(_components.td,{children:_jsx(_components.code,{children:`STANDARD`})})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:`Eight or more messages, with any signal above already present`}),_jsx(_components.td,{children:`whole thread`}),_jsx(_components.td,{children:_jsx(_components.code,{children:`STANDARD`})})]})]})]}),`\n`,_jsxs(_components.p,{children:[`Error detection is not keyword spotting on the word “error”. It matches real failure shapes across languages: tracebacks, panics, segfaults, borrow checker complaints, `,_jsx(_components.code,{children:`npm ERR!`}),`, `,_jsx(_components.code,{children:`ModuleNotFoundError`}),`, `,_jsx(_components.code,{children:`NullPointerException`}),`, `,_jsx(_components.code,{children:`cannot read propert...`}),`, index out of bounds.`]}),`\n`,_jsx(Heading,{level:`3`,id:`what-it-reads`,children:`What it reads`}),`\n`,_jsxs(_components.p,{children:[`The last `,_jsx(_components.strong,{children:`6 conversational messages`}),`, capped at `,_jsx(_components.strong,{children:`4,000 characters each`}),`, system and tool messages excluded. The current prompt is deliberately left out, since it has already been scored on its own merits. Threads shorter than two turns are skipped entirely.`]}),`\n`,_jsx(Note,{children:_jsxs(_components.p,{children:[`History alone never forces `,_jsx(_components.code,{children:`COMPLEX`}),`. The floor is capped at `,_jsx(_components.code,{children:`STANDARD`}),`, so a long conversation can keep you in capable territory but cannot by itself escalate you to the most expensive tier. Reaching `,_jsx(_components.code,{children:`COMPLEX`}),` requires the current prompt to earn it.`]})}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`saying-“thanks”-does-not-cost-you-money`,children:`Saying “thanks” does not cost you money`}),`\n`,_jsxs(_components.p,{children:[`The obvious failure of a floor is that it makes everything expensive forever. Type “hi” in a coding session and you would pay `,_jsx(_components.code,{children:`STANDARD`}),` rates to be told hello.`]}),`\n`,_jsx(_components.p,{children:`So trivial messages are exempt. Anything at or under 60 characters that is recognisably an acknowledgement skips the floor and routes on its own merits:`}),`\n`,_jsx(CodeBlock,{filename:``,numberOfLines:`3`,language:`text`,children:_jsx(_components.pre,{className:`shiki shiki-themes github-light-default dark-plus`,style:{backgroundColor:`#ffffff`,\"--shiki-dark-bg\":`#0B0C0E`,color:`#1f2328`,\"--shiki-dark\":`#D4D4D4`},language:`text`,children:_jsxs(_components.code,{language:`text`,numberOfLines:`3`,children:[_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:`hi hey ok sure thanks thx got it sounds good`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:`cool nice perfect alright yep nope done bye 👍 🙏 🎉`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{})})]})})}),`\n`,_jsx(_components.p,{children:`A greeting in the middle of a long debugging session still goes to a cheap model. The floor comes back the moment you say something real.`}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`category-carry-forward`,children:`Category carry forward`}),`\n`,_jsx(_components.p,{children:`The floor decides how capable the model must be. A second, quieter mechanism decides what kind of model it should be.`}),`\n`,_jsx(_components.p,{children:`When history shows a coding session, Routor remembers the task category and carries it forward through thin follow ups. A vague “now the tests too” stays in the coding category instead of drifting into general chat, because the session already established what you are doing. An error trace marks the session as debugging specifically.`}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`seeing-it-happen`,children:`Seeing it happen`}),`\n`,_jsx(_components.p,{children:`When the floor is applied, the response says so:`}),`\n`,_jsx(CodeBlock,{filename:``,numberOfLines:`4`,language:`text`,children:_jsx(_components.pre,{className:`shiki shiki-themes github-light-default dark-plus`,style:{backgroundColor:`#ffffff`,\"--shiki-dark-bg\":`#0B0C0E`,color:`#1f2328`,\"--shiki-dark\":`#D4D4D4`},language:`text`,children:_jsxs(_components.code,{language:`text`,numberOfLines:`4`,children:[_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:`X-Routor-History-Floor STANDARD`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:`X-Routor-Tier STANDARD`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:`X-Routor-Reasoning history_floor:STANDARD history:error-trace@turn-2 ...`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{})})]})})}),`\n`,_jsx(_components.p,{children:`The reasoning string names each signal and the turn it came from, so you can tell whether it was the stack trace you pasted or the code block the model wrote that held the floor up. When no floor applies, the header is absent.`}),`\n`,_jsxs(_components.p,{children:[`See `,_jsx(_components.a,{href:`/router-metadata`,children:`Router Metadata`}),` for the full header list.`]}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`why-this-matters-more-for-agents`,children:`Why this matters more for agents`}),`\n`,_jsx(_components.p,{children:`Agent loops are where this pays for itself. An agent sends dozens of short, context free instructions: “continue”, “fix it”, “now the tests”. Every one of those, judged alone, looks trivial. Judged in context, most of them are the hard part of a coding task.`}),`\n`,_jsx(_components.p,{children:`Without conversation memory, a per prompt router quietly downgrades an agent at exactly the wrong moments, and the loop produces worse work while appearing to save money.`})]})}function MDXContent(props={}){const{wrapper:MDXLayout}={..._provideComponents(),...props.components};return MDXLayout?_jsx(MDXLayout,{...props,children:_jsx(_createMdxContent,{...props})}):_createMdxContent(props)}return{default:MDXContent};function _missingMdxReference(id,component){throw new Error(`Expected `+(component?`component`:`object`)+\" `\"+id+\"` to be defined: you likely forgot to import, pass, or provide it.\")}","frontmatter":{},"scope":{"config":{},"pageMetadata":{"title":"Conversation Memory","description":"Why a two word follow up in a long debugging session does not get routed to a cheap model.","icon":"comments","href":"/conversation-memory"}}},"mdxSourceWithNoJs":null},"mdxExtracts":{"tableOfContents":[{"title":"The history floor","slug":"the-history-floor","depth":2,"children":[{"title":"What raises it","slug":"what-raises-it","depth":3,"children":[]},{"title":"What it reads","slug":"what-it-reads","depth":3,"children":[]}]},{"title":"Saying “thanks” does not cost you money","slug":"saying-“thanks”-does-not-cost-you-money","depth":2,"children":[]},{"title":"Category carry forward","slug":"category-carry-forward","depth":2,"children":[]},{"title":"Seeing it happen","slug":"seeing-it-happen","depth":2,"children":[]},{"title":"Why this matters more for agents","slug":"why-this-matters-more-for-agents","depth":2,"children":[]}],"codeExamples":{}},"jsLinks":[],"subdomain":"routor","siteName":"Routor","title":"Conversation Memory - Routor","navMaps":{"firstHrefInVersion":{},"firstHrefInLanguage":{}},"lastModified":"2026-09-14T11:22:33.632Z","originalFileLocation":"conversation-memory.md"},"headHtml":"\u003cscript data-mint-head>(function(){var d=document.documentElement,p=null;if(!p){try{p=localStorage.getItem(\"isDarkMode\")}catch(e){}if(p==='true')p='dark';else if(p==='false')p='light';if(p!=='light'\u0026\u0026p!=='dark'\u0026\u0026p!=='system')p=\"system\"}var r=p==='system'?(matchMedia(\"(prefers-color-scheme: dark)\").matches?'dark':'light'):p;d.classList.remove(r==='dark'?'light':'dark');d.classList.add(r);d.setAttribute(\"data-theme-preference\",p)})();\u003c/script>\u003cmeta data-mint-head name=\"description\" content=\"Why a two word follow up in a long debugging session does not get routed to a cheap model.\">\u003cmeta data-mint-head name=\"application-name\" content=\"Routor\">\u003cmeta data-mint-head name=\"generator\" content=\"Mintlify\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"192x192\" href=\"/mintlify-assets/_mintlify/favicons/routor/QowmtP8MueYPVCRM/_generated/favicon/android-chrome-192x192.png\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"apple-touch-icon\" type=\"image/png\" sizes=\"180x180\" href=\"/mintlify-assets/_mintlify/favicons/routor/QowmtP8MueYPVCRM/_generated/favicon/apple-touch-icon.png\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"16x16\" href=\"/mintlify-assets/_mintlify/favicons/routor/QowmtP8MueYPVCRM/_generated/favicon/favicon-16x16.png\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"32x32\" href=\"/mintlify-assets/_mintlify/favicons/routor/QowmtP8MueYPVCRM/_generated/favicon/favicon-32x32.png\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"shortcut icon\" type=\"image/x-icon\" href=\"/mintlify-assets/_mintlify/favicons/routor/QowmtP8MueYPVCRM/_generated/favicon/favicon.ico\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"192x192\" href=\"/mintlify-assets/_mintlify/favicons/routor/QowmtP8MueYPVCRM/_generated/favicon-dark/android-chrome-192x192.png\" media=\"(prefers-color-scheme: dark)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"16x16\" href=\"/mintlify-assets/_mintlify/favicons/routor/QowmtP8MueYPVCRM/_generated/favicon-dark/favicon-16x16.png\" media=\"(prefers-color-scheme: dark)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"32x32\" href=\"/mintlify-assets/_mintlify/favicons/routor/QowmtP8MueYPVCRM/_generated/favicon-dark/favicon-32x32.png\" media=\"(prefers-color-scheme: dark)\">\u003clink data-mint-head rel=\"shortcut icon\" type=\"image/x-icon\" href=\"/mintlify-assets/_mintlify/favicons/routor/QowmtP8MueYPVCRM/_generated/favicon-dark/favicon.ico\" media=\"(prefers-color-scheme: dark)\">\u003cmeta data-mint-head property=\"og:title\" content=\"Conversation Memory - Routor\">\u003cmeta data-mint-head property=\"og:description\" content=\"Why a two word follow up in a long debugging session does not get routed to a cheap model.\">\u003cmeta data-mint-head property=\"og:site_name\" content=\"Routor\">\u003cmeta data-mint-head property=\"og:url\" content=\"https://docs.routor.io/conversation-memory\">\u003cmeta data-mint-head property=\"og:type\" content=\"website\">\u003cmeta data-mint-head property=\"og:image\" content=\"https://routor.mintlify.site/_mintlify/api/og/v2?division=Core+Concepts\u0026amp;title=Conversation+Memory\u0026amp;description=Why+a+two+word+follow+up+in+a+long+debugging+session+does+not+get+routed+to+a+cheap+model.\u0026amp;theme=1416b11e8d88001f9d154536\u0026amp;signature=eab4ede68d9ca354b604c91f1b3467382569bfc98520f126514c109f03a18857\">\u003cmeta data-mint-head property=\"og:image:width\" content=\"1200\">\u003cmeta data-mint-head property=\"og:image:height\" content=\"630\">\u003cmeta data-mint-head name=\"twitter:card\" content=\"summary_large_image\">\u003cmeta data-mint-head name=\"twitter:title\" content=\"Conversation Memory - Routor\">\u003cmeta data-mint-head name=\"twitter:description\" content=\"Why a two word follow up in a long debugging session does not get routed to a cheap model.\">\u003cmeta data-mint-head name=\"twitter:image\" content=\"https://routor.mintlify.site/_mintlify/api/og/v2?division=Core+Concepts\u0026amp;title=Conversation+Memory\u0026amp;description=Why+a+two+word+follow+up+in+a+long+debugging+session+does+not+get+routed+to+a+cheap+model.\u0026amp;theme=1416b11e8d88001f9d154536\u0026amp;signature=eab4ede68d9ca354b604c91f1b3467382569bfc98520f126514c109f03a18857\">\u003cmeta data-mint-head name=\"twitter:image:width\" content=\"1200\">\u003cmeta data-mint-head name=\"twitter:image:height\" content=\"630\">\u003cmeta data-mint-head name=\"msapplication-config\" content=\"/mintlify-assets/_mintlify/favicons/routor/QowmtP8MueYPVCRM/_generated/favicon/browserconfig.xml\">\u003cmeta data-mint-head name=\"apple-mobile-web-app-title\" content=\"Routor\">\u003cmeta data-mint-head name=\"msapplication-TileColor\" content=\"#16A34A\">\u003clink data-mint-head rel=\"canonical\" href=\"https://docs.routor.io/conversation-memory\">\u003clink data-mint-head rel=\"alternate\" type=\"application/xml\" href=\"/sitemap.xml\">\u003clink data-mint-head rel=\"alternate\" type=\"text/markdown\" href=\"/conversation-memory.md\">\u003cscript data-mint-head type=\"application/ld+json\">{\"@context\":\"https://schema.org\",\"@graph\":[{\"@type\":\"Organization\",\"@id\":\"https://docs.routor.io/#organization\",\"name\":\"Routor\",\"url\":\"https://docs.routor.io\",\"logo\":{\"@type\":\"ImageObject\",\"url\":\"https://mintcdn.com/routor/hv7GNqBT8JfNtoVg/logo/logo-light.svg?fit=max\u0026auto=format\u0026n=hv7GNqBT8JfNtoVg\u0026q=85\u0026s=54f11b9d71088b9ca9f3b6214afed989\"}},{\"@type\":\"WebSite\",\"@id\":\"https://docs.routor.io#website\",\"name\":\"Routor\",\"url\":\"https://docs.routor.io\",\"publisher\":{\"@id\":\"https://docs.routor.io/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https://docs.routor.io/conversation-memory#webpage\",\"url\":\"https://docs.routor.io/conversation-memory\",\"name\":\"Conversation Memory\",\"description\":\"Why a two word follow up in a long debugging session does not get routed to a cheap model.\",\"dateModified\":\"2026-09-14T11:22:33.632Z\",\"isPartOf\":{\"@id\":\"https://docs.routor.io#website\"},\"breadcrumb\":{\"@id\":\"https://docs.routor.io/conversation-memory#breadcrumb\"}},{\"@type\":\"BreadcrumbList\",\"@id\":\"https://docs.routor.io/conversation-memory#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Core Concepts\",\"item\":\"https://docs.routor.io/tiers\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Conversation Memory\",\"item\":\"https://docs.routor.io/conversation-memory\"}]},{\"@type\":[\"Article\",\"TechArticle\"],\"@id\":\"https://docs.routor.io/conversation-memory#article\",\"headline\":\"Conversation Memory\",\"name\":\"Conversation Memory\",\"description\":\"Why a two word follow up in a long debugging session does not get routed to a cheap model.\",\"url\":\"https://docs.routor.io/conversation-memory\",\"mainEntityOfPage\":{\"@id\":\"https://docs.routor.io/conversation-memory#webpage\"},\"dateModified\":\"2026-09-14T11:22:33.632Z\",\"publisher\":{\"@id\":\"https://docs.routor.io/#organization\"},\"isPartOf\":{\"@id\":\"https://docs.routor.io#website\"}}]}\u003c/script>\u003cscript data-mint-head type=\"application/ld+json\">{\"@context\":\"https://schema.org\",\"@type\":\"WebSite\",\"name\":\"Routor\",\"publisher\":{\"@type\":\"Organization\",\"name\":\"Routor\"},\"provider\":{\"@type\":\"Organization\",\"name\":\"Mintlify\",\"url\":\"https://mintlify.com\"}}\u003c/script>","agentPointerHref":"/conversation-memory.md","siteName":"Routor","nav":{"global":null,"tabs":[{"tab":"Docs","groups":[{"group":"Getting Started","pages":[{"title":"Introduction","icon":"house","href":"/introduction"},{"title":"The Problem with AI Spend","icon":"triangle-exclamation","href":"/the-problem"},{"title":"Quickstart","icon":"bolt","href":"/quickstart"},{"title":"How Routor Works","icon":"magnifying-glass","href":"/how-it-works"},{"title":"Architecture and Vision","icon":"diagram-project","href":"/architecture"},{"title":"Plans \u0026 Pricing","icon":"credit-card","href":"/plans-pricing"},{"title":"FAQ","icon":"circle-question","href":"/faq"}]},{"group":"Core Concepts","pages":[{"title":"The 5 Routing Tiers","icon":"layer-group","href":"/tiers"},{"title":"Why Not Just Auto?","description":"How Routor decides which model answers, and why that differs from routers that follow the market.","icon":"route","href":"/why-not-just-auto"},{"title":"Conversation Memory","icon":"comments","href":"/conversation-memory"},{"title":"Router Metadata","description":"Every response tells you which model answered, why it was chosen, and what it saved.","icon":"tags","href":"/router-metadata"},{"title":"Fallback and Auto-Failover","icon":"shield-halved","href":"/fallback"}]},{"group":"Ecosystem","pages":[{"title":"$ROUTOR Ecosystem","icon":"coins","href":"/routor-token"},{"title":"Holder Rewards","icon":"gift","href":"/ecosystem/holder-rewards"},{"title":"The $ROUTOR Flywheel","icon":"arrows-rotate","href":"/ecosystem/flywheel"}]},{"group":"Dashboard","pages":[{"title":"Dashboard Overview","icon":"chart-line","href":"/dashboard/overview"},{"title":"API Keys","icon":"key","href":"/dashboard/api-keys"},{"title":"Billing and Credits","icon":"credit-card","href":"/dashboard/billing"},{"title":"Providers","icon":"circle-nodes","href":"/dashboard/providers"},{"title":"Models","icon":"layer-group","href":"/dashboard/models"}]},{"group":"Playground","pages":[{"title":"What is the Playground","icon":"square-terminal","href":"/playground/overview"},{"title":"Creating a Routing Profile","icon":"sliders","href":"/playground/create-profile"},{"title":"Testing Prompts","icon":"vial","href":"/playground/testing"},{"title":"Customer-Facing Chatbot","icon":"comments","href":"/use-cases/chatbot"},{"title":"Code Assistant","icon":"code","href":"/use-cases/code-assistant"},{"title":"Customer Support Automation","icon":"headset","href":"/use-cases/customer-support"}]},{"group":"Providers","pages":[{"title":"Supported Providers and Models","icon":"circle-nodes","href":"/providers/supported"},{"title":"Provider Health Monitoring","icon":"heart-pulse","href":"/providers/health"}]},{"group":"Guides","pages":[{"title":"Build Your First App with Routor in Node.js","icon":"node-js","href":"/guides/nodejs-first-app"},{"title":"Add Image Understanding to Your App","icon":"image","href":"/guides/vision"},{"title":"Stream Responses in Your App","icon":"bolt","href":"/guides/streaming"},{"title":"Use Tool Calling with Routor","icon":"wrench","href":"/guides/tool-calling"},{"title":"Migrate from OpenAI SDK","icon":"arrow-right-arrow-left","href":"/guides/migrate-openai"},{"title":"Migrate from Anthropic SDK","icon":"arrow-right-arrow-left","href":"/guides/migrate-anthropic"},{"title":"LangChain and LangGraph","icon":"link","href":"/guides/langchain"},{"title":"Use with Claude Code","icon":"terminal","href":"/guides/claude-code"},{"title":"Use with Cline CLI","icon":"terminal","href":"/guides/cline"}]}]},{"tab":"API Reference","pages":[{"title":"API Reference","href":"/api/chat-completions"}]},{"tab":"Future","pages":[{"title":"Future","href":"/future/overview"}]},{"tab":"Roadmap","pages":[{"title":"Roadmap","href":"/roadmap"}]},{"tab":"Support","pages":[{"title":"Support","href":"/troubleshooting"}]}]},"basePath":"","title":"Conversation Memory - Routor","deploymentId":"dpl_DyXGnKeCjZyA3sUdfMJ7FjkdBefg","buildId":"dpl_DyXGnKeCjZyA3sUdfMJ7FjkdBefg","dynamics":[],"chunks":[]}