Claude Haiku 5.5: Guenstiger Preis gilt nur bis 100.000 Tokens
Anthropic hat Claude Haiku 5.5 veroeffentlicht, die Modell-ID lautet claude-haiku-5-5. Bis 100.000 Tokens kostet der Input 0,10 Dollar pro Million Tokens und der Output 0,50 Dollar, rund 90 Prozent weniger als beim Vorgaenger. Darueber hinaus steigen die Preise auf 0,50 beziehungsweise 2,50 Dollar.
Am 7. Oktober 2026 hat Anthropic Claude Haiku 5.5 veroeffentlicht. Die Modell-ID lautet claude-haiku-5-5. Fuer Prompts bis 100.000 Token kostet der Input 0,10 US-Dollar pro Million Token, der Output 0,50 US-Dollar. Das liegt 90 Prozent unter den Preisen von Haiku 4.5 mit 1,00 und 5,00 US-Dollar. Ueber 100.000 Token hinaus gelten die Saetze 0,50 und 2,50 US-Dollar. Das ist eine Reduzierung von 50 Prozent, nicht von 90 Prozent.
Anthropic gibt zudem an, dass der neue Tokenizer fuer denselben Text etwa 30 Prozent mehr Token benoetigt als Haiku 4.5. Ein guenstigerer Token bedeutet nicht automatisch eine guenstigere Aufgabe. Und Haiku 5.5 ist das erste Haiku-Modell mit einem Effort-Regler. Der API-Standardwert ist medium. low, high, xhigh und max sind echte Einstellungen. Sie veraendern, wie viel das Modell nachdenkt, und das veraendert die Rechnung. Der Listenpreis verraet das nicht.
Die nuetzliche Frage lautet, was Ihr Code tut, bevor die Anfrage den Rechner verlaesst.
Dieses kleine Werkzeug ruft Claude nicht auf. Man uebergibt ihm bereits vorhandene Token-Zahlen, einen Aufgabennamen und einen Effort-Wert. Es antwortet mit ALLOW, REVIEW oder REFUSE und gibt die Rechnung zum Listenpreis in US-Dollar aus.
Die Prompt-Groesse ist die Summe aus Input-Token, Cache-Lesezugriffen und Cache-Schreibzugriffen. Bei genau 100.000 oder weniger gelten die kurzen Saetze. Ueber dieser Grenze gelten die langen Saetze, und die Entscheidung lautet REVIEW.
ALLOW gilt nur fuer fuenf enge Aufgaben: classify, extract, route, summarize und compact. Der Effort-Wert muss low oder medium sein. Laesst man Effort weg, stempelt der Gate medium ein, denn das ist der API-Standardwert fuer dieses Modell.
agentic-coding und computer-use fuehren zu REVIEW, selbst bei einem kurzen Prompt. Anthropics eigene Terminal-Bench-4.0-Zahlen liegen bei 39,2 Prozent fuer Haiku 5.5 und 70,6 Prozent fuer Sonnet 5.5. Ein niedrigerer Preis ist nicht dieselbe Aufgabe. Auch high, xhigh und max fuehren zu REVIEW. Dieses Programm hat keinen Eval, der belegt, dass sich diese Stufen rechnen.
Falsche Zaehlungen und unbekannte Bezeichnungen fuehren zu REFUSE. Eine negative Token-Zahl sollte nicht zu einem Rabatt werden.
Der vollstaendige Quellcode folgt unten. Speichern Sie ihn als src/gate.ts. Node 22 oder neuer. Keine Pakete. Kein Schluessel.
node --experimental-strip-types src/gate.ts Enter fullscreen mode Exit fullscreen mode /** * Price gate for Claude Haiku 5.5. * No network. No API key. Rates are list prices in microdollars per million tokens. * * Sources, read 2026-10-08: * - https://www.anthropic.com/claude-haiku-5-5 (October 7, 2026) * - https://platform.claude.com/docs/en/models/haiku-5-5/overview * - https://platform.claude.com/docs/en/build-with-claude/effort */ const MILLION = 1 _000_000n; const PROMPT_LIMIT = 100 _000n; const RECOUNT_NUM = 130 n; const RECOUNT_DEN = 100 n; const ALLOW_TASKS = new Set ([ " classify ", " extract ", " route ", " summarize ", " compact ", ]); const REVIEW_TASKS = new Set ([ " agentic-coding ", " computer-use " ]); const ALLOW_EFFORT = new Set ([ " low ", " medium " ]); const ALL_EFFORT = new Set ([ " low ", " medium ", " high ", " xhigh ", " max " ]); type Effort = " low " | " medium " | " high " | " xhigh " | " max "; type Ttl = " 5m " | " 1h "; type Tier = " short " | " long "; type Rates = { input: bigint; output: bigint; cacheRead: bigint; cacheWrite5m: bigint; cacheWrite1h: bigint; }; const HAIKU_55: Record < Tier, Rates > = { short: { input: 100 _000n, output: 500 _000n, cacheRead: 10 _000n, cacheWrite5m: 125 _000n, cacheWrite1h: 200 _000n, }, long: { input: 500 _000n, output: 2 _500_000n, cacheRead: 50 _000n, cacheWrite5m: 625 _000n, cacheWrite1h: 1 _000_000n, }, }; const HAIKU_45_5M: Rates = { input: 1 _000_000n, output: 5 _000_000n, cacheRead: 100 _000n, cacheWrite5m: 1 _250_000n, cacheWrite1h: 0 n, }; const SONNET_55_CACHE_READ_NEW = 100 _000n; const SONNET_55_CACHE_READ_OLD = 200 _000n; export type Job = { name: string; task: string; effort?: string; inputTokens: number; outputTokens: number; cacheReadTokens: number; cacheWriteTokens: number; cacheWriteTtl?: Ttl; recountFromHaiku45?: boolean; }; export type Decision = " ALLOW " | " REVIEW " | " REFUSE "; export type Quote = { name: string; decision: Decision; reason: string; model: " claude-haiku-5-5 "; effort: Effort | null; assumedDefaultEffort: boolean; tier: Tier | null; promptTokens: bigint | null; haiku55Usd: string | null; haiku45Usd: string | null; savingsBps: number | null; sonnetCacheReadNewUsd: string | null; sonnetCacheReadOldUsd: string | null; }; function isWhole ( value: number ): boolean { return Number. isFinite ( value ) && Number. isInteger ( value ) && value >= 0; } function mulDiv ( tokens: bigint, perMillion: bigint ): bigint { return ( tokens * perMillion ) / MILLION; } function usd ( micro: bigint ): string { const whole = micro / MILLION; const frac = ( micro % MILLION ). toString (). padStart ( 6, " 0 " ); return ` ${ whole }. ${ frac } `; } function bill ( tokens: { input: bigint; output: bigint; cacheRead: bigint; cacheWrite: bigint; }, rates: Rates, ttl: Ttl ): bigint { const writeRate = ttl === " 1h "? rates. cacheWrite1h: rates. cacheWrite5m; return ( mulDiv ( tokens. input, rates. input ) + mulDiv ( tokens. output, rates. output ) + mulDiv ( tokens. cacheRead, rates. cacheRead ) + mulDiv ( tokens. cacheWrite, writeRate ) ); } function savingsBps ( before: bigint, after: bigint ): number | null { if ( before <= 0 n ) return null; return Number ((( before - after ) * 10 _000n ) / before ); } export function quoteJob ( job: Job ): Quote { const base = { name: job. name, model: " claude-haiku-5-5 " as const, effort: null, assumedDefaultEffort: false, tier: null, promptTokens: null, haiku55Usd: null, haiku45Usd: null, savingsBps: null, sonnetCacheReadNewUsd: null, sonnetCacheReadOldUsd: null, }; const counts = [ job. inputTokens, job. outputTokens, job. cacheReadTokens, job. cacheWriteTokens, ]; if (! counts. every ( isWhole )) { return {... base, decision: " REFUSE ", reason: " token counts must be non-negative integers ", }; } const ttl = job. cacheWriteTtl??
" 5m "; if ( ttl!== " 5m " && ttl!== " 1h " ) { return {... base, decision: " REFUSE ", reason: " cache write ttl must be 5m or 1h " }; } let effort = job. effort; let assumedDefaultEffort = false; if ( effort === undefined ) { effort = " medium "; assumedDefaultEffort = true; } if (! ALL_EFFORT. has ( effort )) { return {... base, decision: " REFUSE ", reason: `unknown effort ${ job. effort } ` }; } const knownTask = ALLOW_TASKS. has ( job. task ) || REVIEW_TASKS. has ( job. task ); if (! knownTask ) { return {... base, decision: " REFUSE ", reason: `unknown task ${ job. task } ` }; } let input = BigInt ( job. inputTokens ); let cacheRead = BigInt ( job. cacheReadTokens ); let cacheWrite = BigInt ( job. cacheWriteTokens ); const output = BigInt ( job. outputTokens ); if ( job. recountFromHaiku45 ) { input = ( input * RECOUNT_NUM ) / RECOUNT_DEN; cacheRead = ( cacheRead * RECOUNT_NUM ) / RECOUNT_DEN; cacheWrite = ( cacheWrite * RECOUNT_NUM ) / RECOUNT_DEN; } const promptTokens = input + cacheRead + cacheWrite; const tier: Tier = promptTokens <= PROMPT_LIMIT? " short ": " long "; const haiku55 = bill ( { input, output, cacheRead, cacheWrite }, HAIKU_55 [ tier ], ttl, ); let haiku45: bigint | null = null; if ( ttl === " 5m " ) { const oldInput = BigInt ( job. inputTokens ); const oldRead = BigInt ( job. cacheReadTokens ); const oldWrite = BigInt ( job. cacheWriteTokens ); haiku45 = bill ( { input: oldInput, output, cacheRead: oldRead, cacheWrite: oldWrite }, HAIKU_45_5M, " 5m ", ); } const reasons: string [] = []; if (!
ALLOW_TASKS. has ( job. task )) { reasons. push ( ` ${ job. task } bleibt von der Haiku-Allow-Liste ausgeschlossen` ); } if (! ALLOW_EFFORT. has ( effort )) { reasons. push ( `effort ${ effort } benoetigt einen Eval, den dieses Werkzeug nicht hat` ); } if ( tier === " long " ) { reasons. push ( " der Prompt liegt ueber 100000 Token, daher gelten die Saetze der langen Stufe " ); } if ( job. recountFromHaiku45 ) { reasons. push ( " die Prompt-Token wurden mit 130/100 neu gezaehlt, bevor die Haiku-5.5-Rechnung erstellt wurde " ); } if ( assumedDefaultEffort ) { reasons. push ( " der fehlende Effort-Wert wurde mit medium eingestempelt, dem API-Standard " ); } const decision: Decision = reasons. some (( reason ) => reason. startsWith ( " prompt is over " ) || reason. includes ( " allow-list " ) || reason. includes ( " needs an eval " ) )? " REVIEW ": " ALLOW "; if ( decision === " ALLOW " ) { reasons. unshift ( " kurzer Prompt, Aufgabe auf der Allow-Liste, Effort ist low oder medium " ); } if ( ttl === " 1h " ) { reasons. push ( " Ein-Stunden-Cache-Schreibzugriffe werden nur fuer Haiku 5.5 ausgewiesen " ); } const sonnetNew = mulDiv ( cacheRead, SONNET_55_CACHE_READ_NEW ); const sonnetOld = mulDiv ( BigInt ( job. cacheReadTokens ), SONNET_55_CACHE_READ_OLD ); return {... base, decision, reason: reasons. join ( "; " ), effort: effort as Effort, assumedDefaultEffort, tier, promptTokens, haiku55Usd: usd ( haiku55 ), haiku45Usd: haiku45 === null? null: usd ( haiku45 ), savingsBps: haiku45 === null? null: savingsBps ( haiku45, haiku55 ), sonnetCacheReadNewUsd: usd ( sonnetNew ), sonnetCacheReadOldUsd: usd ( sonnetOld ), }; } function line ( quote: Quote ): string { const parts = [ quote. name, quote. decision, quote. model, quote. effort === null?
" effort=none ": `effort= ${ quote. effort } `, quote. tier === null? " tier=none ": `tier= ${ quote. tier } `, quote. promptTokens === null? " prompt=none ": `prompt= ${ quote. promptTokens } `, quote. haiku55Usd === null?
" haiku55=none ": `haiku55= ${ quote. haiku55Usd } `, quote. haiku45Usd === null? " haiku45=none ": `haiku45= ${ quote. haiku45Usd } `, quote. savingsBps === null? " savings_bps=none ": `savings_bps= ${ quote. savingsBps } `, quote. reason, ]; return parts. join ( " | " ); } const fixtures: Job [] = [ { name: " classify-short ", task: " classify ", effort: " medium ", inputTokens: 2 _000, outputTokens: 200, cacheReadTokens: 0, cacheWriteTokens: 0, }, { name: " compact-under-limit ", task: " compact ", effort: " low ", inputTokens: 4 _000, outputTokens: 500, cacheReadTokens: 80 _000, cacheWriteTokens: 0, }, { name: " compact-over-limit ", task: " compact ", effort: " medium ", inputTokens: 20 _000, outputTokens: 400, cacheReadTokens: 90 _000, cacheWriteTokens: 0, }, { name: " coding-short ", task: " agentic-coding ", effort: " medium ", inputTokens: 8 _000, outputTokens: 1 _200, cacheReadTokens: 0, cacheWriteTokens: 0, }, { name: " classify-max-effort ", task: " classify ", effort: " max ", inputTokens: 2 _000, outputTokens: 200, cacheReadTokens: 0, cacheWriteTokens: 0, }, { name: " extract-default-effort ", task: " extract ", inputTokens: 1 _500, outputTokens: 300, cacheReadTokens: 0, cacheWriteTokens: 0, }, { name: " recount-30 ", task: " summarize ", effort: " medium ", inputTokens: 10 _000, outputTokens: 500, cacheReadTokens: 0, cacheWriteTokens: 0, recountFromHaiku45: true, }, { name: " one-hour-write ", task: " route ", effort: " low ", inputTokens: 1 _000, outputTokens: 100, cacheReadTokens: 0, cacheWriteTokens: 4 _000, cacheWriteTtl: " 1h ", }, { name: " bad-tokens ", task: " classify ", effort: " low ", inputTokens: - 1, outputTokens: 10, cacheReadTokens: 0, cacheWriteTokens: 0, }, { name: " unknown-effort ", task: " classify ", effort: " turbo ", inputTokens: 100, outputTokens: 10, cacheReadTokens: 0, cacheWriteTokens: 0, }, ]; const expected: Record < string, { decision: Decision; haiku55Usd: string | null; savingsBps: number | null } > = { " classify-short ": { decision: " ALLOW ", haiku55Usd: " 0.000300 ", savingsBps: 9000 }, " compact-under-limit ": { decision: " ALLOW ", haiku55Usd: " 0.001450 ", savingsBps: 9000 }, " compact-over-limit ": { decision: " REVIEW ", haiku55Usd: " 0.015500 ", savingsBps: 5000 }, " coding-short ": { decision: " REVIEW ", haiku55Usd: " 0.001400 ", savingsBps: 9000 }, " classify-max-effort ": { decision: " REVIEW ", haiku55Usd: " 0.000300 ", savingsBps: 9000 }, " extract-default-effort ": { decision: " ALLOW ", haiku55Usd: " 0.000300 ", savingsBps: 9000 }, " recount-30 ": { decision: " ALLOW ", haiku55Usd: " 0.001550 ", savingsBps: 8760 }, " one-hour-write ": { decision: " ALLOW ", haiku55Usd: " 0.000950 ", savingsBps: null }, " bad-tokens ": { decision: " REFUSE ", haiku55Usd: null, savingsBps: null }, " unknown-effort ": { decision: " REFUSE ", haiku55Usd: null, savingsBps: null }, }; function main (): void { let failed = 0; for ( const job of fixtures ) { const quote = quoteJob ( job ); const want = expected [ job. name ]; const ok = want. decision === quote. decision && want. haiku55Usd === quote. haiku55Usd && want. savingsBps === quote. savingsBps; if (! ok ) { failed += 1; console. error ( `FAIL ${ job. name } ` ); console. error ( quote ); } console. log ( line ( quote )); } const cacheRead = 100 _000n; const sonnetNew = usd ( mulDiv ( cacheRead, SONNET_55_CACHE_READ_NEW )); const sonnetOld = usd ( mulDiv ( cacheRead, SONNET_55_CACHE_READ_OLD )); console. log ( `sonnet-cache-read | QUOTE | claude-sonnet-5-5 | cache_read_tokens=100000 | new= ${ sonnetNew } | old= ${ sonnetOld } | October 7 cache-read cut, not a routing decision`, ); if ( sonnetNew!== " 0.010000 " || sonnetOld!== " 0.020000 " ) { failed += 1; console. error ( " FAIL sonnet-cache-read " ); } if ( failed > 0 ) { console. error ( ` ${ failed } check(s) failed` ); process. exit ( 1 ); } console. log ( `checks=11 passed` ); } main (); Vollbildmodus aktivieren Vollbildmodus beenden Geprüft am 8.
Oktober 2026, Node.js 25.6.0.
Elf Prüfungen bestanden:
classify-short | ALLOW | claude-haiku -5-5 | effort=medium | tier=short | prompt= 2000 | haiku 55 = 0.000300 | haiku 45 = 0.003000 | savings_bps= 9000 | kurzer Prompt, allow-gelistete Aufgabe, effort ist low oder medium compact-under-limit | ALLOW | claude-haiku -5-5 | effort=low | tier=short | prompt= 84000 | haiku 55 = 0.001450 | haiku 45 = 0.014500 | savings_bps= 9000 | kurzer Prompt, allow-gelistete Aufgabe, effort ist low oder medium compact-over-limit | REVIEW | claude-haiku -5-5 | effort=medium | tier=long | prompt= 110000 | haiku 55 = 0.015500 | haiku 45 = 0.031000 | savings_bps= 5000 | der Prompt liegt über 100000 Token, daher gelten die Long-Tier-Sätze coding-short | REVIEW | claude-haiku -5-5 | effort=medium | tier=short | prompt= 8000 | haiku 55 = 0.001400 | haiku 45 = 0.014000 | savings_bps= 9000 | agentic-coding bleibt von der Haiku-Allow-Liste ausgeschlossen classify-max-effort | REVIEW | claude-haiku -5-5 | effort=max | tier=short | prompt= 2000 | haiku 55 = 0.000300 | haiku 45 = 0.003000 | savings_bps= 9000 | effort max erfordert einen Eval, den dieses Spielzeugmodell nicht hat extract-default-effort | ALLOW | claude-haiku -5-5 | effort=medium | tier=short | prompt= 1500 | haiku 55 = 0.000300 | haiku 45 = 0.003000 | savings_bps= 9000 | kurzer Prompt, allow-gelistete Aufgabe, effort ist low oder medium; der fehlende effort wurde mit medium, dem API-Standardwert, versehen recount -30 | ALLOW | claude-haiku -5-5 | effort=medium | tier=short | prompt= 13000 | haiku 55 = 0.001550 | haiku 45 = 0.012500 | savings_bps= 8760 | kurzer Prompt, allow-gelistete Aufgabe, effort ist low oder medium; die Prompt-Token wurden vor der Haiku-5.5-Abrechnung mit 130/100 neu gezählt one-hour-write | ALLOW | claude-haiku -5-5 | effort=low | tier=short | prompt= 5000 | haiku 55 = 0.000950 | haiku 45 =none | savings_bps=none | kurzer Prompt, allow-gelistete Aufgabe, effort ist low oder medium; Cache-Writes mit einer Stunde werden nur für Haiku 5.5 quotiert bad-tokens | REFUSE | claude-haiku -5-5 | effort=none | tier=none | prompt=none | haiku 55 =none | haiku 45 =none | savings_bps=none | die Token-Zahlen müssen nichtnegative ganze Zahlen sein unknown-effort | REFUSE | claude-haiku -5-5 | effort=none | tier=none | prompt=none | haiku 55 =none | haiku 45 =none | savings_bps=none | unbekannter effort turbo sonnet-cache-read | QUOTE | claude-sonnet -5-5 | cache_read_tokens= 100000 | new= 0.010000 | old= 0.020000 | Senkung der Cache-Read-Preise am 7. Oktober, keine Routing-Entscheidung checks= 11 passed Vollbildmodus aktivieren Vollbildmodus beenden savings_bps sind Basispunkte gegenüber einer Haiku-4.5-Abrechnung zu den ursprünglichen Token-Zahlen. 9000 entspricht 90,00 %.
5000 entspricht 50,00 %.
8760 entspricht 87,60 %.
classify-short ist der Musterfall. 2.000 Input-Token und 200 Output-Token, effort medium. Haiku 5.5 kostet 0,000300 Dollar. Haiku 4.5 kostet 0,003000 Dollar. Gleiche Token, 90 Prozent weniger.
compact-over-limit ist die Grenze, die man überschauen kann. 20.000 frische Input-Token plus 90.000 Cache-Read-Token ergeben einen Prompt mit 110.000 Token. Der Auftrag ist weiterhin compact und damit auf der Allow-Liste, effort ist weiterhin medium. Die Entscheidung lautet REVIEW. Haiku 5.5 kostet 0,015500 Dollar. Haiku 4.5 kostet 0,031000 Dollar. Die Ersparnis beträgt 50 Prozent, denn die Long-Tier-Sätze sind fünfmal so hoch wie die Short-Tier-Sätze, und der Pausalsatz von Haiku 4.5 liegt zehnmal über dem Short-Tier-Satz.
recount-30 geht von einer Haiku-4.5-Zählung mit 10.000 Input-Token und 500 Output-Token aus. Das Gate multipliziert die Input-Token mit 130/100, bevor es Haiku 5.5 bepreist, daher basiert die neue Abrechnung auf 13.000 Input-Token. Haiku 5.5 kostet 0,001550 Dollar. Haiku 4.5 kostet 0,012500 Dollar. Das sind 87,60 Prozent weniger, nicht 90 Prozent.
Das letzte Angebot ist kein Routing. Am 7. Oktober senkte Anthropic die Kosten für Sonnet-5.5-Cache-Reads von 0,20 auf 0,10 Dollar pro Million Token. 100.000 Cache-Read-Token kosten jetzt 0,010000 Dollar, zuvor waren es 0,020000 Dollar. Das Gate gibt das aus. Es schickt den Auftrag aber nicht an Sonnet.
Ich baue Roster, KI-Mitarbeiter, die echte Arbeit leisten. Auch ein kleineres Modell braucht ein Gate, das weiß, welchen Preis es gleich zahlen wird.
Mit Vorlagen lassen sich FAQs schnell beantworten oder Snippets zur Wiederverwendung speichern.
Sind Sie sicher, dass Sie diesen Kommentar verbergen möchten? Er wird in Ihrem Beitrag nicht mehr angezeigt, bleibt aber über den Permalink des Kommentars sichtbar.
Für weitere Maßnahmen können Sie diese Person blockieren und/oder einen Missbrauchsfall melden.
Automatisch übersetzt. Das Original auf Englisch ist unten aufklappbar.
Original (Englisch) anzeigen
On October 7, 2026, Anthropic released Claude Haiku 5.5. The model id is claude-haiku-5-5. For prompts up to 100,000 tokens, input is $0.10 per million tokens and output is $0.50. That is 90% below Haiku 4.5's $1.00 and $5.00. Cross 100,000 tokens and the rates become $0.50 and $2.50. That is a 50% cut, not a 90% cut.
Anthropic also says the new tokenizer uses about 30% more tokens than Haiku 4.5 for the same text. A cheaper token is not automatically a cheaper task. And Haiku 5.5 is the first Haiku with an effort dial. The API default is medium. low, high, xhigh, and max are real settings. They change how much the model thinks, which changes the bill. The list price does not tell you that.
The useful question is what your code does before the request leaves.
This toy does not call Claude. You hand it token counts you already have, a task name, and an effort. It answers ALLOW, REVIEW, or REFUSE, and it prints the list-price bill in USD.
Prompt size is input tokens plus cache reads plus cache writes. At or under 100,000, the short rates apply. Over that line, the long rates apply and the decision is REVIEW.
ALLOW is only for five narrow jobs: classify, extract, route, summarize, and compact. Effort must be low or medium. If you omit effort, the gate stamps medium, because that is the Claude API default for this model.
agentic-coding and computer-use are REVIEW even on a short prompt. Anthropic's own Terminal-Bench 4.0 numbers are 39.2% for Haiku 5.5 and 70.6% for Sonnet 5.5. A lower price is not the same job. high, xhigh, and max are also REVIEW. This program has no eval that says those levels pay for themselves.
Bad counts and unknown labels are REFUSE. A negative token count should not become a discount.
The full source is below. Save it as src/gate.ts. Node 22 or newer. No packages. No key.
node --experimental-strip-types src/gate.ts Enter fullscreen mode Exit fullscreen mode /** * Price gate for Claude Haiku 5.5. * No network. No API key. Rates are list prices in microdollars per million tokens. * * Sources, read 2026-10-08: * - https://www.anthropic.com/claude-haiku-5-5 (October 7, 2026) * - https://platform.claude.com/docs/en/models/haiku-5-5/overview * - https://platform.claude.com/docs/en/build-with-claude/effort */ const MILLION = 1 _000_000n; const PROMPT_LIMIT = 100 _000n; const RECOUNT_NUM = 130 n; const RECOUNT_DEN = 100 n; const ALLOW_TASKS = new Set ([ " classify ", " extract ", " route ", " summarize ", " compact ", ]); const REVIEW_TASKS = new Set ([ " agentic-coding ", " computer-use " ]); const ALLOW_EFFORT = new Set ([ " low ", " medium " ]); const ALL_EFFORT = new Set ([ " low ", " medium ", " high ", " xhigh ", " max " ]); type Effort = " low " | " medium " | " high " | " xhigh " | " max "; type Ttl = " 5m " | " 1h "; type Tier = " short " | " long "; type Rates = { input: bigint; output: bigint; cacheRead: bigint; cacheWrite5m: bigint; cacheWrite1h: bigint; }; const HAIKU_55: Record < Tier, Rates > = { short: { input: 100 _000n, output: 500 _000n, cacheRead: 10 _000n, cacheWrite5m: 125 _000n, cacheWrite1h: 200 _000n, }, long: { input: 500 _000n, output: 2 _500_000n, cacheRead: 50 _000n, cacheWrite5m: 625 _000n, cacheWrite1h: 1 _000_000n, }, }; const HAIKU_45_5M: Rates = { input: 1 _000_000n, output: 5 _000_000n, cacheRead: 100 _000n, cacheWrite5m: 1 _250_000n, cacheWrite1h: 0 n, }; const SONNET_55_CACHE_READ_NEW = 100 _000n; const SONNET_55_CACHE_READ_OLD = 200 _000n; export type Job = { name: string; task: string; effort?: string; inputTokens: number; outputTokens: number; cacheReadTokens: number; cacheWriteTokens: number; cacheWriteTtl?: Ttl; recountFromHaiku45?: boolean; }; export type Decision = " ALLOW " | " REVIEW " | " REFUSE "; export type Quote = { name: string; decision: Decision; reason: string; model: " claude-haiku-5-5 "; effort: Effort | null; assumedDefaultEffort: boolean; tier: Tier | null; promptTokens: bigint | null; haiku55Usd: string | null; haiku45Usd: string | null; savingsBps: number | null; sonnetCacheReadNewUsd: string | null; sonnetCacheReadOldUsd: string | null; }; function isWhole ( value: number ): boolean { return Number. isFinite ( value ) && Number. isInteger ( value ) && value >= 0; } function mulDiv ( tokens: bigint, perMillion: bigint ): bigint { return ( tokens * perMillion ) / MILLION; } function usd ( micro: bigint ): string { const whole = micro / MILLION; const frac = ( micro % MILLION ). toString (). padStart ( 6, " 0 " ); return ` ${ whole }. ${ frac } `; } function bill ( tokens: { input: bigint; output: bigint; cacheRead: bigint; cacheWrite: bigint; }, rates: Rates, ttl: Ttl ): bigint { const writeRate = ttl === " 1h "? rates. cacheWrite1h: rates. cacheWrite5m; return ( mulDiv ( tokens. input, rates. input ) + mulDiv ( tokens. output, rates. output ) + mulDiv ( tokens. cacheRead, rates. cacheRead ) + mulDiv ( tokens. cacheWrite, writeRate ) ); } function savingsBps ( before: bigint, after: bigint ): number | null { if ( before <= 0 n ) return null; return Number ((( before - after ) * 10 _000n ) / before ); } export function quoteJob ( job: Job ): Quote { const base = { name: job. name, model: " claude-haiku-5-5 " as const, effort: null, assumedDefaultEffort: false, tier: null, promptTokens: null, haiku55Usd: null, haiku45Usd: null, savingsBps: null, sonnetCacheReadNewUsd: null, sonnetCacheReadOldUsd: null, }; const counts = [ job. inputTokens, job. outputTokens, job. cacheReadTokens, job. cacheWriteTokens, ]; if (! counts. every ( isWhole )) { return {... base, decision: " REFUSE ", reason: " token counts must be non-negative integers ", }; } const ttl = job. cacheWriteTtl??
" 5m "; if ( ttl!== " 5m " && ttl!== " 1h " ) { return {... base, decision: " REFUSE ", reason: " cache write ttl must be 5m or 1h " }; } let effort = job. effort; let assumedDefaultEffort = false; if ( effort === undefined ) { effort = " medium "; assumedDefaultEffort = true; } if (! ALL_EFFORT. has ( effort )) { return {... base, decision: " REFUSE ", reason: `unknown effort ${ job. effort } ` }; } const knownTask = ALLOW_TASKS. has ( job. task ) || REVIEW_TASKS. has ( job. task ); if (! knownTask ) { return {... base, decision: " REFUSE ", reason: `unknown task ${ job. task } ` }; } let input = BigInt ( job. inputTokens ); let cacheRead = BigInt ( job. cacheReadTokens ); let cacheWrite = BigInt ( job. cacheWriteTokens ); const output = BigInt ( job. outputTokens ); if ( job. recountFromHaiku45 ) { input = ( input * RECOUNT_NUM ) / RECOUNT_DEN; cacheRead = ( cacheRead * RECOUNT_NUM ) / RECOUNT_DEN; cacheWrite = ( cacheWrite * RECOUNT_NUM ) / RECOUNT_DEN; } const promptTokens = input + cacheRead + cacheWrite; const tier: Tier = promptTokens <= PROMPT_LIMIT? " short ": " long "; const haiku55 = bill ( { input, output, cacheRead, cacheWrite }, HAIKU_55 [ tier ], ttl, ); let haiku45: bigint | null = null; if ( ttl === " 5m " ) { const oldInput = BigInt ( job. inputTokens ); const oldRead = BigInt ( job. cacheReadTokens ); const oldWrite = BigInt ( job. cacheWriteTokens ); haiku45 = bill ( { input: oldInput, output, cacheRead: oldRead, cacheWrite: oldWrite }, HAIKU_45_5M, " 5m ", ); } const reasons: string [] = []; if (!
ALLOW_TASKS. has ( job. task )) { reasons. push ( ` ${ job. task } stays off the Haiku allow-list` ); } if (! ALLOW_EFFORT. has ( effort )) { reasons. push ( `effort ${ effort } needs an eval this toy does not have` ); } if ( tier === " long " ) { reasons. push ( " prompt is over 100000 tokens, so the long-tier rates apply " ); } if ( job. recountFromHaiku45 ) { reasons. push ( " prompt tokens were recounted at 130/100 before the Haiku 5.5 bill " ); } if ( assumedDefaultEffort ) { reasons. push ( " missing effort was stamped medium, the API default " ); } const decision: Decision = reasons. some (( reason ) => reason. startsWith ( " prompt is over " ) || reason. includes ( " allow-list " ) || reason. includes ( " needs an eval " ) )? " REVIEW ": " ALLOW "; if ( decision === " ALLOW " ) { reasons. unshift ( " short prompt, allow-listed task, effort is low or medium " ); } if ( ttl === " 1h " ) { reasons. push ( " one-hour cache writes are quoted for Haiku 5.5 only " ); } const sonnetNew = mulDiv ( cacheRead, SONNET_55_CACHE_READ_NEW ); const sonnetOld = mulDiv ( BigInt ( job. cacheReadTokens ), SONNET_55_CACHE_READ_OLD ); return {... base, decision, reason: reasons. join ( "; " ), effort: effort as Effort, assumedDefaultEffort, tier, promptTokens, haiku55Usd: usd ( haiku55 ), haiku45Usd: haiku45 === null? null: usd ( haiku45 ), savingsBps: haiku45 === null? null: savingsBps ( haiku45, haiku55 ), sonnetCacheReadNewUsd: usd ( sonnetNew ), sonnetCacheReadOldUsd: usd ( sonnetOld ), }; } function line ( quote: Quote ): string { const parts = [ quote. name, quote. decision, quote. model, quote. effort === null?
" effort=none ": `effort= ${ quote. effort } `, quote. tier === null? " tier=none ": `tier= ${ quote. tier } `, quote. promptTokens === null? " prompt=none ": `prompt= ${ quote. promptTokens } `, quote. haiku55Usd === null?
" haiku55=none ": `haiku55= ${ quote. haiku55Usd } `, quote. haiku45Usd === null? " haiku45=none ": `haiku45= ${ quote. haiku45Usd } `, quote. savingsBps === null? " savings_bps=none ": `savings_bps= ${ quote. savingsBps } `, quote. reason, ]; return parts. join ( " | " ); } const fixtures: Job [] = [ { name: " classify-short ", task: " classify ", effort: " medium ", inputTokens: 2 _000, outputTokens: 200, cacheReadTokens: 0, cacheWriteTokens: 0, }, { name: " compact-under-limit ", task: " compact ", effort: " low ", inputTokens: 4 _000, outputTokens: 500, cacheReadTokens: 80 _000, cacheWriteTokens: 0, }, { name: " compact-over-limit ", task: " compact ", effort: " medium ", inputTokens: 20 _000, outputTokens: 400, cacheReadTokens: 90 _000, cacheWriteTokens: 0, }, { name: " coding-short ", task: " agentic-coding ", effort: " medium ", inputTokens: 8 _000, outputTokens: 1 _200, cacheReadTokens: 0, cacheWriteTokens: 0, }, { name: " classify-max-effort ", task: " classify ", effort: " max ", inputTokens: 2 _000, outputTokens: 200, cacheReadTokens: 0, cacheWriteTokens: 0, }, { name: " extract-default-effort ", task: " extract ", inputTokens: 1 _500, outputTokens: 300, cacheReadTokens: 0, cacheWriteTokens: 0, }, { name: " recount-30 ", task: " summarize ", effort: " medium ", inputTokens: 10 _000, outputTokens: 500, cacheReadTokens: 0, cacheWriteTokens: 0, recountFromHaiku45: true, }, { name: " one-hour-write ", task: " route ", effort: " low ", inputTokens: 1 _000, outputTokens: 100, cacheReadTokens: 0, cacheWriteTokens: 4 _000, cacheWriteTtl: " 1h ", }, { name: " bad-tokens ", task: " classify ", effort: " low ", inputTokens: - 1, outputTokens: 10, cacheReadTokens: 0, cacheWriteTokens: 0, }, { name: " unknown-effort ", task: " classify ", effort: " turbo ", inputTokens: 100, outputTokens: 10, cacheReadTokens: 0, cacheWriteTokens: 0, }, ]; const expected: Record < string, { decision: Decision; haiku55Usd: string | null; savingsBps: number | null } > = { " classify-short ": { decision: " ALLOW ", haiku55Usd: " 0.000300 ", savingsBps: 9000 }, " compact-under-limit ": { decision: " ALLOW ", haiku55Usd: " 0.001450 ", savingsBps: 9000 }, " compact-over-limit ": { decision: " REVIEW ", haiku55Usd: " 0.015500 ", savingsBps: 5000 }, " coding-short ": { decision: " REVIEW ", haiku55Usd: " 0.001400 ", savingsBps: 9000 }, " classify-max-effort ": { decision: " REVIEW ", haiku55Usd: " 0.000300 ", savingsBps: 9000 }, " extract-default-effort ": { decision: " ALLOW ", haiku55Usd: " 0.000300 ", savingsBps: 9000 }, " recount-30 ": { decision: " ALLOW ", haiku55Usd: " 0.001550 ", savingsBps: 8760 }, " one-hour-write ": { decision: " ALLOW ", haiku55Usd: " 0.000950 ", savingsBps: null }, " bad-tokens ": { decision: " REFUSE ", haiku55Usd: null, savingsBps: null }, " unknown-effort ": { decision: " REFUSE ", haiku55Usd: null, savingsBps: null }, }; function main (): void { let failed = 0; for ( const job of fixtures ) { const quote = quoteJob ( job ); const want = expected [ job. name ]; const ok = want. decision === quote. decision && want. haiku55Usd === quote. haiku55Usd && want. savingsBps === quote. savingsBps; if (! ok ) { failed += 1; console. error ( `FAIL ${ job. name } ` ); console. error ( quote ); } console. log ( line ( quote )); } const cacheRead = 100 _000n; const sonnetNew = usd ( mulDiv ( cacheRead, SONNET_55_CACHE_READ_NEW )); const sonnetOld = usd ( mulDiv ( cacheRead, SONNET_55_CACHE_READ_OLD )); console. log ( `sonnet-cache-read | QUOTE | claude-sonnet-5-5 | cache_read_tokens=100000 | new= ${ sonnetNew } | old= ${ sonnetOld } | October 7 cache-read cut, not a routing decision`, ); if ( sonnetNew!== " 0.010000 " || sonnetOld!== " 0.020000 " ) { failed += 1; console. error ( " FAIL sonnet-cache-read " ); } if ( failed > 0 ) { console. error ( ` ${ failed } check(s) failed` ); process. exit ( 1 ); } console. log ( `checks=11 passed` ); } main (); Enter fullscreen mode Exit fullscreen mode Checked on October 8, 2026, Node.js 25.6.0.
Eleven checks passed:
classify-short | ALLOW | claude-haiku -5-5 | effort=medium | tier=short | prompt= 2000 | haiku 55 = 0.000300 | haiku 45 = 0.003000 | savings_bps= 9000 | short prompt, allow-listed task, effort is low or medium compact-under-limit | ALLOW | claude-haiku -5-5 | effort=low | tier=short | prompt= 84000 | haiku 55 = 0.001450 | haiku 45 = 0.014500 | savings_bps= 9000 | short prompt, allow-listed task, effort is low or medium compact-over-limit | REVIEW | claude-haiku -5-5 | effort=medium | tier=long | prompt= 110000 | haiku 55 = 0.015500 | haiku 45 = 0.031000 | savings_bps= 5000 | prompt is over 100000 tokens, so the long-tier rates apply coding-short | REVIEW | claude-haiku -5-5 | effort=medium | tier=short | prompt= 8000 | haiku 55 = 0.001400 | haiku 45 = 0.014000 | savings_bps= 9000 | agentic-coding stays off the Haiku allow-list classify-max-effort | REVIEW | claude-haiku -5-5 | effort=max | tier=short | prompt= 2000 | haiku 55 = 0.000300 | haiku 45 = 0.003000 | savings_bps= 9000 | effort max needs an eval this toy does not have extract-default-effort | ALLOW | claude-haiku -5-5 | effort=medium | tier=short | prompt= 1500 | haiku 55 = 0.000300 | haiku 45 = 0.003000 | savings_bps= 9000 | short prompt, allow-listed task, effort is low or medium; missing effort was stamped medium, the API default recount -30 | ALLOW | claude-haiku -5-5 | effort=medium | tier=short | prompt= 13000 | haiku 55 = 0.001550 | haiku 45 = 0.012500 | savings_bps= 8760 | short prompt, allow-listed task, effort is low or medium; prompt tokens were recounted at 130 / 100 before the Haiku 5.5 bill one-hour-write | ALLOW | claude-haiku -5-5 | effort=low | tier=short | prompt= 5000 | haiku 55 = 0.000950 | haiku 45 =none | savings_bps=none | short prompt, allow-listed task, effort is low or medium; one-hour cache writes are quoted for Haiku 5.5 only bad-tokens | REFUSE | claude-haiku -5-5 | effort=none | tier=none | prompt=none | haiku 55 =none | haiku 45 =none | savings_bps=none | token counts must be non-negative integers unknown-effort | REFUSE | claude-haiku -5-5 | effort=none | tier=none | prompt=none | haiku 55 =none | haiku 45 =none | savings_bps=none | unknown effort turbo sonnet-cache-read | QUOTE | claude-sonnet -5-5 | cache_read_tokens= 100000 | new= 0.010000 | old= 0.020000 | October 7 cache-read cut, not a routing decision checks= 11 passed Enter fullscreen mode Exit fullscreen mode savings_bps is basis points against a Haiku 4.5 bill at the original token counts. 9000 is 90.00%. 5000 is 50.00%.
8760 is 87.60%.
classify-short is the brochure case. 2,000 input tokens and 200 output tokens, effort medium. Haiku 5.5 is $0.000300. Haiku 4.5 is $0.003000. Same tokens, 90% less.
compact-over-limit is the line you can miss. 20,000 fresh input tokens plus 90,000 cache-read tokens is a 110,000-token prompt. The job is still compact, which is on the allow-list, and effort is still medium. The decision is REVIEW. Haiku 5.5 costs $0.015500. Haiku 4.5 costs $0.031000. The saving is 50%, because the long-tier rates are five times the short-tier rates, and Haiku 4.5's flat rate is ten times the short-tier rate.
recount-30 starts from a Haiku 4.5 count of 10,000 input tokens and 500 output tokens. The gate multiplies the input by 130/100 before it prices Haiku 5.5, so the new bill uses 13,000 input tokens. Haiku 5.5 is $0.001550. Haiku 4.5 is $0.012500. That is 87.60% less, not 90%.
The last quote is not a route. On October 7, Anthropic cut Sonnet 5.5 cache reads from $0.20 to $0.10 per million tokens. 100,000 cache-read tokens are $0.010000 now and were $0.020000. The gate prints that. It does not send the job to Sonnet.
I'm building Roster, AI employees that do real work. A smaller model still needs a gate that knows which price it is about to pay.