{"id":10286,"date":"2026-08-15T01:19:40","date_gmt":"2026-08-14T23:19:40","guid":{"rendered":"https:\/\/www.lukaswojcik.com\/blog\/toolbox\/rag-chunking-embedding-planner\/"},"modified":"2026-08-15T01:19:40","modified_gmt":"2026-08-14T23:19:40","slug":"rag-chunking-embedding-planner","status":"publish","type":"page","link":"https:\/\/www.lukaswojcik.com\/blog\/en\/toolbox\/rag-chunking-embedding-planner\/","title":{"rendered":"RAG Chunking &#038; Embedding Planner"},"content":{"rendered":"<div class=\"gtm-analyser-container\" style=\"background: var(--bg-panel, #1e1e24); padding: 25px; border-radius: 8px; border: 1px solid var(--border, #2a2a35);\">\n<p style=\"color: var(--text-secondary, #a0a0b0); margin-bottom: 20px;\">Chunk size and overlap decide three things at once: how many chunks exist, what the embedding run costs, and how much of the context window a single answer consumes. This planner shows all three before the first document is indexed.<\/p>\n<div style=\"display: flex; gap: 12px; flex-wrap: wrap;\">\n<div style=\"flex: 1; min-width: 150px;\">\n            <label for=\"rg-dokumente\" style=\"color: var(--text-secondary, #a0a0b0); display: block; font-size: 0.85rem; margin-bottom: 5px;\">Documents<\/label><br \/>\n            <input type=\"text\" id=\"rg-dokumente\" class=\"form-control\" style=\"width: 100%; padding: 10px; background: var(--bg-body, #14141a); border: 1px solid var(--border, #2a2a35); color: var(--text-primary, #e8e8ee); border-radius: 6px; font-family: monospace; box-sizing: border-box;\" value=\"2500\">\n        <\/div>\n<div style=\"flex: 1; min-width: 150px;\">\n            <label for=\"rg-woerter\" style=\"color: var(--text-secondary, #a0a0b0); display: block; font-size: 0.85rem; margin-bottom: 5px;\">Words per document<\/label><br \/>\n            <input type=\"text\" id=\"rg-woerter\" class=\"form-control\" style=\"width: 100%; padding: 10px; background: var(--bg-body, #14141a); border: 1px solid var(--border, #2a2a35); color: var(--text-primary, #e8e8ee); border-radius: 6px; font-family: monospace; box-sizing: border-box;\" value=\"900\">\n        <\/div>\n<div style=\"flex: 1; min-width: 150px;\">\n            <label for=\"rg-chunk\" style=\"color: var(--text-secondary, #a0a0b0); display: block; font-size: 0.85rem; margin-bottom: 5px;\">Chunk size in tokens<\/label><br \/>\n            <input type=\"text\" id=\"rg-chunk\" class=\"form-control\" style=\"width: 100%; padding: 10px; background: var(--bg-body, #14141a); border: 1px solid var(--border, #2a2a35); color: var(--text-primary, #e8e8ee); border-radius: 6px; font-family: monospace; box-sizing: border-box;\" value=\"512\">\n        <\/div>\n<div style=\"flex: 1; min-width: 150px;\">\n            <label for=\"rg-ueberlappung\" style=\"color: var(--text-secondary, #a0a0b0); display: block; font-size: 0.85rem; margin-bottom: 5px;\">Overlap in tokens<\/label><br \/>\n            <input type=\"text\" id=\"rg-ueberlappung\" class=\"form-control\" style=\"width: 100%; padding: 10px; background: var(--bg-body, #14141a); border: 1px solid var(--border, #2a2a35); color: var(--text-primary, #e8e8ee); border-radius: 6px; font-family: monospace; box-sizing: border-box;\" value=\"64\">\n        <\/div>\n<\/p><\/div>\n<div style=\"display: flex; gap: 12px; flex-wrap: wrap; margin-top: 12px;\">\n<div style=\"flex: 1; min-width: 150px;\">\n            <label for=\"rg-dimensionen\" style=\"color: var(--text-secondary, #a0a0b0); display: block; font-size: 0.85rem; margin-bottom: 5px;\">Embedding dimensions<\/label><br \/>\n            <input type=\"text\" id=\"rg-dimensionen\" class=\"form-control\" style=\"width: 100%; padding: 10px; background: var(--bg-body, #14141a); border: 1px solid var(--border, #2a2a35); color: var(--text-primary, #e8e8ee); border-radius: 6px; font-family: monospace; box-sizing: border-box;\" value=\"768\">\n        <\/div>\n<div style=\"flex: 1; min-width: 150px;\">\n            <label for=\"rg-preis\" style=\"color: var(--text-secondary, #a0a0b0); display: block; font-size: 0.85rem; margin-bottom: 5px;\">Price per million tokens<\/label><br \/>\n            <input type=\"text\" id=\"rg-preis\" class=\"form-control\" style=\"width: 100%; padding: 10px; background: var(--bg-body, #14141a); border: 1px solid var(--border, #2a2a35); color: var(--text-primary, #e8e8ee); border-radius: 6px; font-family: monospace; box-sizing: border-box;\" value=\"0.13\">\n        <\/div>\n<div style=\"flex: 1; min-width: 150px;\">\n            <label for=\"rg-treffer\" style=\"color: var(--text-secondary, #a0a0b0); display: block; font-size: 0.85rem; margin-bottom: 5px;\">Chunks retrieved per query<\/label><br \/>\n            <input type=\"text\" id=\"rg-treffer\" class=\"form-control\" style=\"width: 100%; padding: 10px; background: var(--bg-body, #14141a); border: 1px solid var(--border, #2a2a35); color: var(--text-primary, #e8e8ee); border-radius: 6px; font-family: monospace; box-sizing: border-box;\" value=\"5\">\n        <\/div>\n<div style=\"flex: 1; min-width: 150px;\">\n            <label for=\"rg-fenster\" style=\"color: var(--text-secondary, #a0a0b0); display: block; font-size: 0.85rem; margin-bottom: 5px;\">Context window in tokens<\/label><br \/>\n            <input type=\"text\" id=\"rg-fenster\" class=\"form-control\" style=\"width: 100%; padding: 10px; background: var(--bg-body, #14141a); border: 1px solid var(--border, #2a2a35); color: var(--text-primary, #e8e8ee); border-radius: 6px; font-family: monospace; box-sizing: border-box;\" value=\"8000\">\n        <\/div>\n<\/p><\/div>\n<div style=\"display: flex; gap: 10px; flex-wrap: wrap; margin: 18px 0 0;\">\n        <button id=\"rg-btn\" class=\"button\" style=\"background: var(--accent, #7ee787); color: #0b1114; border: none; padding: 12px 24px; border-radius: 6px; font-weight: 700; cursor: pointer;\">Plan it<\/button>\n    <\/div>\n<div id=\"rg-ergebnis\" style=\"margin-top: 22px;\"><\/div>\n<\/div>\n<p><script>\n(function () {\n    'use strict';<\/p>\n<p>    const T = {\"einleitung\":\"Chunk size and overlap decide three things at once: how many chunks exist, what the embedding run costs, and how much of the context window a single answer consumes. This planner shows all three before the first document is indexed.\",\"label_dokumente\":\"Documents\",\"label_woerter\":\"Words per document\",\"label_chunk\":\"Chunk size in tokens\",\"label_ueberlappung\":\"Overlap in tokens\",\"label_dimensionen\":\"Embedding dimensions\",\"label_preis\":\"Price per million tokens\",\"label_treffer\":\"Chunks retrieved per query\",\"label_fenster\":\"Context window in tokens\",\"knopf\":\"Plan it\",\"fehler_dokumente\":\"The number of documents has to be above zero.\",\"fehler_woerter\":\"The words per document have to be above zero.\",\"fehler_chunk\":\"The chunk size has to be above zero.\",\"fehler_ueberlappung\":\"The overlap has to be a number of 0 or more.\",\"fehler_ueberlappung_gross\":\"The overlap has to stay below the chunk size, otherwise the chunks never move forward.\",\"fehler_dimensionen\":\"The dimensions have to be above zero.\",\"fehler_preis\":\"The price has to be a number of 0 or more.\",\"fehler_treffer\":\"At least one chunk has to be retrieved.\",\"fehler_fenster\":\"The context window has to be above zero.\",\"h_zerlegung\":\"How it splits\",\"h_kosten\":\"One-off cost and storage\",\"h_abfrage\":\"Per query\",\"h_befunde\":\"Findings\",\"z_je_dokument\":\"Chunks per document\",\"z_chunks\":\"Chunks in total\",\"z_tokens_netto\":\"Tokens in the documents\",\"z_tokens_gesamt\":\"Tokens to embed, overlap included\",\"z_aufschlag\":\"Surcharge from the overlap\",\"z_kosten\":\"Embedding run\",\"z_vektoren\":\"Vectors\",\"z_speicher\":\"Vectors and text together\",\"z_kontext\":\"Retrieved per query\",\"z_kontext_anteil\":\"Share of the context window\",\"b_ueberlappung\":\"An overlap of %s percent means %s percent more tokens to embed and to store, for the same content.\",\"b_ohne_ueberlappung\":\"Without overlap a sentence that straddles a boundary is torn apart, and neither half carries the whole statement.\",\"b_kontext_zu_gross\":\"%k tokens of retrieved text do not fit into a window of %f. Either fewer chunks or smaller ones.\",\"b_kontext_eng\":\"The retrieved chunks already fill %s percent of the window. What is left has to carry the system prompt, the question and the answer.\",\"b_chunk_gross\":\"Chunks above a thousand tokens carry several topics at once, and the embedding averages over all of them. That is what makes retrieval vague.\",\"b_chunk_klein\":\"Chunks below two hundred tokens rarely hold a complete thought, so the retrieval finds the right words without the context around them.\",\"b_ein_chunk\":\"The documents fit into a single chunk each. Then the split does nothing and the whole document is the unit of retrieval.\",\"b_viele_chunks\":\"Above a million vectors an exact search gets slow. That is where an approximate index and its parameters start to matter.\",\"b_neu_einbetten\":\"A change of embedding model means running the whole set again, another %s. Model and dimensions are not a detail to decide casually.\",\"b_speicher\":\"The figure covers vectors and raw text. An index on top of them, metadata and copies come in addition, so plan with roughly twice the amount.\"};<\/p>\n<p>    \/\/ --- reine Rechenlogik ---<\/p>\n<p>    function zahl(wert) {\n        const roh = String(wert === undefined || wert === null ? '' : wert).trim().replace(\/\\s\/g, '').replace(',', '.');\n        if (roh === '') { return NaN; }\n        if (!\/^[0-9]+(\\.[0-9]+)?$\/.test(roh)) { return NaN; }\n        return parseFloat(roh);\n    }<\/p>\n<p>    function runde(wert, stellen) {\n        const faktor = Math.pow(10, stellen === undefined ? 2 : stellen);\n        return Math.round(wert * faktor) \/ faktor;\n    }<\/p>\n<p>    function groesse(bytes) {\n        if (bytes < 1024) { return runde(bytes, 0) + ' B'; }\n        if (bytes < 1024 * 1024) { return runde(bytes \/ 1024, 1) + ' KB'; }\n        if (bytes < 1024 * 1024 * 1024) { return runde(bytes \/ 1024 \/ 1024, 1) + ' MB'; }\n        return runde(bytes \/ 1024 \/ 1024 \/ 1024, 2) + ' GB';\n    }\n\n    function planen(e) {\n        const ergebnis = { ok: false, fehler: '', befunde: [] };\n        const eingabe = e || {};\n\n        const dokumente = zahl(eingabe.dokumente);\n        const woerter = zahl(eingabe.woerter);\n        const chunk = zahl(eingabe.chunk);\n        const ueberlappung = zahl(eingabe.ueberlappung);\n        const dimensionen = zahl(eingabe.dimensionen);\n        const preis = zahl(eingabe.preis);\n        const treffer = zahl(eingabe.treffer);\n        const fenster = zahl(eingabe.fenster);\n\n        if (!isFinite(dokumente) || dokumente <= 0) { ergebnis.fehler = T.fehler_dokumente; return ergebnis; }\n        if (!isFinite(woerter) || woerter <= 0) { ergebnis.fehler = T.fehler_woerter; return ergebnis; }\n        if (!isFinite(chunk) || chunk <= 0) { ergebnis.fehler = T.fehler_chunk; return ergebnis; }\n        if (!isFinite(ueberlappung) || ueberlappung < 0) { ergebnis.fehler = T.fehler_ueberlappung; return ergebnis; }\n        if (ueberlappung >= chunk) { ergebnis.fehler = T.fehler_ueberlappung_gross; return ergebnis; }\n        if (!isFinite(dimensionen) || dimensionen <= 0) { ergebnis.fehler = T.fehler_dimensionen; return ergebnis; }\n        if (!isFinite(preis) || preis < 0) { ergebnis.fehler = T.fehler_preis; return ergebnis; }\n        if (!isFinite(treffer) || treffer < 1) { ergebnis.fehler = T.fehler_treffer; return ergebnis; }\n        if (!isFinite(fenster) || fenster <= 0) { ergebnis.fehler = T.fehler_fenster; return ergebnis; }\n\n        \/\/ Ein Wort ergibt im Mittel etwa anderthalb Tokens.\n        const tokensJeDokument = woerter * 1.5;\n        const schritt = chunk - ueberlappung;\n        const chunksJeDokument = Math.max(1, Math.ceil((tokensJeDokument - ueberlappung) \/ schritt));\n        const chunks = chunksJeDokument * dokumente;\n        \/\/ Jeder Chunk traegt seine Ueberlappung noch einmal mit.\n        const tokensGesamt = chunks * chunk;\n        const tokensNetto = tokensJeDokument * dokumente;\n\n        const speicherVektoren = chunks * dimensionen * 4;\n        const speicherText = tokensGesamt * 4;\n        const kontext = treffer * chunk;\n\n        ergebnis.ok = true;\n        ergebnis.chunksJeDokument = chunksJeDokument;\n        ergebnis.chunks = chunks;\n        ergebnis.tokensNetto = Math.round(tokensNetto);\n        ergebnis.tokensGesamt = Math.round(tokensGesamt);\n        ergebnis.aufschlag = runde((tokensGesamt \/ tokensNetto - 1) * 100, 1);\n        ergebnis.kosten = runde(tokensGesamt \/ 1000000 * preis, 2);\n        ergebnis.speicherVektoren = speicherVektoren;\n        ergebnis.speicherText = speicherText;\n        ergebnis.speicherGesamt = groesse(speicherVektoren + speicherText);\n        ergebnis.vektorenText = groesse(speicherVektoren);\n        ergebnis.kontext = kontext;\n        ergebnis.kontextAnteil = runde(kontext \/ fenster * 100, 1);\n\n        \/\/ --- Befunde ---\n        if (ueberlappung \/ chunk > 0.3) {\n            ergebnis.befunde.push({ schwer: true, text: T.b_ueberlappung.replace('%s', String(runde(ueberlappung \/ chunk * 100, 0))).replace('%a', String(ergebnis.aufschlag)) });\n        }\n        if (ueberlappung === 0) {\n            ergebnis.befunde.push({ schwer: true, text: T.b_ohne_ueberlappung });\n        }\n        if (kontext > fenster) {\n            ergebnis.befunde.push({ schwer: true, text: T.b_kontext_zu_gross.replace('%k', String(kontext)).replace('%f', String(fenster)) });\n        } else if (ergebnis.kontextAnteil > 60) {\n            ergebnis.befunde.push({ schwer: true, text: T.b_kontext_eng.replace('%s', String(ergebnis.kontextAnteil)) });\n        }\n        if (chunk > 1000) {\n            ergebnis.befunde.push({ schwer: true, text: T.b_chunk_gross });\n        }\n        if (chunk < 200) {\n            ergebnis.befunde.push({ schwer: true, text: T.b_chunk_klein });\n        }\n        if (chunksJeDokument === 1) {\n            ergebnis.befunde.push({ schwer: false, text: T.b_ein_chunk });\n        }\n        if (chunks > 1000000) {\n            ergebnis.befunde.push({ schwer: true, text: T.b_viele_chunks });\n        }\n        ergebnis.befunde.push({ schwer: false, text: T.b_neu_einbetten.replace('%s', String(ergebnis.kosten)) });\n        ergebnis.befunde.push({ schwer: false, text: T.b_speicher });<\/p>\n<p>        return ergebnis;\n    }<\/p>\n<p>    if (typeof window !== 'undefined') {\n        window.LW_TEST = window.LW_TEST || {};\n        window.LW_TEST.RG = { planen: planen, zahl: zahl, groesse: groesse };\n    }<\/p>\n<p>    \/\/ --- Anbindung an das Dokument ---<\/p>\n<p>    const btn = document.getElementById('rg-btn');\n    const out = document.getElementById('rg-ergebnis');\n    if (!btn || !out) { return; }<\/p>\n<p>    const KASTEN = 'background: var(--bg-body, #14141a); border: 1px solid var(--border, #2a2a35); border-radius: 6px; padding: 14px; margin-bottom: 18px;';\n    const UEBERSCHRIFT = 'color: var(--text-primary, #e8e8ee); font-weight: 700; margin-bottom: 8px;';<\/p>\n<p>    function el(tag, css, inhalt) {\n        const knoten = document.createElement(tag);\n        if (css) { knoten.setAttribute('style', css); }\n        if (inhalt !== undefined && inhalt !== null) { knoten.textContent = inhalt; }\n        return knoten;\n    }<\/p>\n<p>    function wert(id) {\n        const knoten = document.getElementById(id);\n        return knoten ? knoten.value : '';\n    }<\/p>\n<p>    function zeile(kasten, name, inhalt, hervor) {\n        const z = el('div', 'display: flex; justify-content: space-between; gap: 14px; flex-wrap: wrap; padding: 5px 0;');\n        z.appendChild(el('span', 'color: var(--text-secondary, #a0a0b0);', name));\n        z.appendChild(el('span', 'color: ' + (hervor ? 'var(--accent, #7ee787)' : 'var(--text-primary, #e8e8ee)') + '; font-family: monospace;', inhalt));\n        kasten.appendChild(z);\n    }<\/p>\n<p>    btn.addEventListener('click', function () {\n        const erg = planen({\n            dokumente: wert('rg-dokumente'), woerter: wert('rg-woerter'), chunk: wert('rg-chunk'),\n            ueberlappung: wert('rg-ueberlappung'), dimensionen: wert('rg-dimensionen'),\n            preis: wert('rg-preis'), treffer: wert('rg-treffer'), fenster: wert('rg-fenster')\n        });<\/p>\n<p>        out.textContent = '';\n        if (!erg.ok) {\n            out.appendChild(el('div', 'background: var(--bg-body, #14141a); border: 1px solid var(--border, #2a2a35); border-left: 3px solid #ffb86c; color: var(--text-secondary, #a0a0b0); padding: 14px; border-radius: 6px;', erg.fehler));\n            return;\n        }<\/p>\n<p>        out.appendChild(el('div', UEBERSCHRIFT, T.h_zerlegung));\n        const kasten = el('div', KASTEN);\n        zeile(kasten, T.z_je_dokument, String(erg.chunksJeDokument));\n        zeile(kasten, T.z_chunks, String(erg.chunks), true);\n        zeile(kasten, T.z_tokens_netto, String(erg.tokensNetto));\n        zeile(kasten, T.z_tokens_gesamt, String(erg.tokensGesamt));\n        zeile(kasten, T.z_aufschlag, erg.aufschlag + ' %', erg.aufschlag > 30);\n        out.appendChild(kasten);<\/p>\n<p>        out.appendChild(el('div', UEBERSCHRIFT, T.h_kosten));\n        const kosten = el('div', KASTEN);\n        zeile(kosten, T.z_kosten, erg.kosten.toFixed(2), true);\n        zeile(kosten, T.z_vektoren, erg.vektorenText);\n        zeile(kosten, T.z_speicher, erg.speicherGesamt);\n        out.appendChild(kosten);<\/p>\n<p>        out.appendChild(el('div', UEBERSCHRIFT, T.h_abfrage));\n        const abfrage = el('div', KASTEN);\n        zeile(abfrage, T.z_kontext, String(erg.kontext) + ' Tokens');\n        zeile(abfrage, T.z_kontext_anteil, erg.kontextAnteil + ' %', erg.kontextAnteil > 60);\n        out.appendChild(abfrage);<\/p>\n<p>        out.appendChild(el('div', UEBERSCHRIFT, T.h_befunde));\n        const befunde = el('div', KASTEN);\n        erg.befunde.forEach(function (b) {\n            const z = el('div', 'display: flex; gap: 10px; padding: 5px 0;');\n            z.appendChild(el('span', 'color: ' + (b.schwer ? '#ffb86c' : 'var(--text-secondary, #a0a0b0)') + '; font-family: monospace;', b.schwer ? '!' : '-'));\n            z.appendChild(el('span', 'color: var(--text-secondary, #a0a0b0); font-size: 0.9rem; line-height: 1.5;', b.text));\n            befunde.appendChild(z);\n        });\n        out.appendChild(befunde);\n    });\n})();\n<\/script><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Turns a document set into chunk counts, embedding cost and vector storage, and shows how much of the context window the retrieved chunks take up.<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":38,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"template-tool-base.php","meta":{"footnotes":""},"tags":[91333,91407,91404],"class_list":["post-10286","page","type-page","status-publish","hentry","tag-bigquery","tag-google-cloud","tag-llm"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.1 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>RAG Chunking &amp; Embedding Planner - Lukas Wojcik - Blog<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.lukaswojcik.com\/blog\/en\/toolbox\/rag-chunking-embedding-planner\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"RAG Chunking &amp; Embedding Planner - Lukas Wojcik - Blog\" \/>\n<meta property=\"og:description\" content=\"Turns a document set into chunk counts, embedding cost and vector storage, and shows how much of the context window the retrieved chunks take up.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.lukaswojcik.com\/blog\/en\/toolbox\/rag-chunking-embedding-planner\/\" \/>\n<meta property=\"og:site_name\" content=\"Lukas Wojcik - Blog\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.lukaswojcik.com\/blog\/wp-content\/uploads\/2026\/08\/og-default.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1200\" \/>\n\t<meta property=\"og:image:height\" content=\"630\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/en\\\/toolbox\\\/rag-chunking-embedding-planner\\\/\",\"url\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/en\\\/toolbox\\\/rag-chunking-embedding-planner\\\/\",\"name\":\"RAG Chunking & Embedding Planner - Lukas Wojcik - Blog\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/#website\"},\"datePublished\":\"2026-08-14T23:19:40+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/en\\\/toolbox\\\/rag-chunking-embedding-planner\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/en\\\/toolbox\\\/rag-chunking-embedding-planner\\\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/en\\\/toolbox\\\/rag-chunking-embedding-planner\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Toolbox\",\"item\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/en\\\/toolbox\\\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"RAG Chunking &#038; Embedding Planner\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/#website\",\"url\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/\",\"name\":\"Lukas Wojcik - Blog\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/#\\\/schema\\\/person\\\/895f7604f9b6b71aad9bba33af28d0f9\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":[\"Person\",\"Organization\"],\"@id\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/#\\\/schema\\\/person\\\/895f7604f9b6b71aad9bba33af28d0f9\",\"name\":\"luky\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/lw-x2.jpg\",\"url\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/lw-x2.jpg\",\"contentUrl\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/lw-x2.jpg\",\"width\":424,\"height\":636,\"caption\":\"luky\"},\"logo\":{\"@id\":\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/lw-x2.jpg\"},\"sameAs\":[\"https:\\\/\\\/www.lukaswojcik.com\\\/blog\"]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"RAG Chunking & Embedding Planner - Lukas Wojcik - Blog","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.lukaswojcik.com\/blog\/en\/toolbox\/rag-chunking-embedding-planner\/","og_locale":"en_US","og_type":"article","og_title":"RAG Chunking & Embedding Planner - Lukas Wojcik - Blog","og_description":"Turns a document set into chunk counts, embedding cost and vector storage, and shows how much of the context window the retrieved chunks take up.","og_url":"https:\/\/www.lukaswojcik.com\/blog\/en\/toolbox\/rag-chunking-embedding-planner\/","og_site_name":"Lukas Wojcik - Blog","og_image":[{"width":1200,"height":630,"url":"https:\/\/www.lukaswojcik.com\/blog\/wp-content\/uploads\/2026\/08\/og-default.jpg","type":"image\/jpeg"}],"twitter_card":"summary_large_image","schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.lukaswojcik.com\/blog\/en\/toolbox\/rag-chunking-embedding-planner\/","url":"https:\/\/www.lukaswojcik.com\/blog\/en\/toolbox\/rag-chunking-embedding-planner\/","name":"RAG Chunking & Embedding Planner - Lukas Wojcik - Blog","isPartOf":{"@id":"https:\/\/www.lukaswojcik.com\/blog\/#website"},"datePublished":"2026-08-14T23:19:40+00:00","breadcrumb":{"@id":"https:\/\/www.lukaswojcik.com\/blog\/en\/toolbox\/rag-chunking-embedding-planner\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.lukaswojcik.com\/blog\/en\/toolbox\/rag-chunking-embedding-planner\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.lukaswojcik.com\/blog\/en\/toolbox\/rag-chunking-embedding-planner\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.lukaswojcik.com\/blog\/"},{"@type":"ListItem","position":2,"name":"Toolbox","item":"https:\/\/www.lukaswojcik.com\/blog\/en\/toolbox\/"},{"@type":"ListItem","position":3,"name":"RAG Chunking &#038; Embedding Planner"}]},{"@type":"WebSite","@id":"https:\/\/www.lukaswojcik.com\/blog\/#website","url":"https:\/\/www.lukaswojcik.com\/blog\/","name":"Lukas Wojcik - Blog","description":"","publisher":{"@id":"https:\/\/www.lukaswojcik.com\/blog\/#\/schema\/person\/895f7604f9b6b71aad9bba33af28d0f9"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.lukaswojcik.com\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":["Person","Organization"],"@id":"https:\/\/www.lukaswojcik.com\/blog\/#\/schema\/person\/895f7604f9b6b71aad9bba33af28d0f9","name":"luky","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.lukaswojcik.com\/blog\/wp-content\/uploads\/2026\/07\/lw-x2.jpg","url":"https:\/\/www.lukaswojcik.com\/blog\/wp-content\/uploads\/2026\/07\/lw-x2.jpg","contentUrl":"https:\/\/www.lukaswojcik.com\/blog\/wp-content\/uploads\/2026\/07\/lw-x2.jpg","width":424,"height":636,"caption":"luky"},"logo":{"@id":"https:\/\/www.lukaswojcik.com\/blog\/wp-content\/uploads\/2026\/07\/lw-x2.jpg"},"sameAs":["https:\/\/www.lukaswojcik.com\/blog"]}]}},"_links":{"self":[{"href":"https:\/\/www.lukaswojcik.com\/blog\/wp-json\/wp\/v2\/pages\/10286","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.lukaswojcik.com\/blog\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.lukaswojcik.com\/blog\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.lukaswojcik.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.lukaswojcik.com\/blog\/wp-json\/wp\/v2\/comments?post=10286"}],"version-history":[{"count":0,"href":"https:\/\/www.lukaswojcik.com\/blog\/wp-json\/wp\/v2\/pages\/10286\/revisions"}],"up":[{"embeddable":true,"href":"https:\/\/www.lukaswojcik.com\/blog\/wp-json\/wp\/v2\/pages\/38"}],"wp:attachment":[{"href":"https:\/\/www.lukaswojcik.com\/blog\/wp-json\/wp\/v2\/media?parent=10286"}],"wp:term":[{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.lukaswojcik.com\/blog\/wp-json\/wp\/v2\/tags?post=10286"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}