The document processing quality of Knowledge Bases (KBSphere) is determined by these settings.
Properly configuring content extraction engines, text splitting, and embedding models is essential for RAG search accuracy.
Manage all RAG pipeline settings in Admin > Settings > Documents
Extraction Engine
In the Extraction Engine section at the top of the page, register and manage the engine instances used to extract document text.
Extraction is not “picking a single engine globally” — instead, you register named instances and select them per extension within a profile.
1
Register an engine instance
Add an instance with the + button in the Extraction Engine section, then specify a Name and an Engine Type. The selectable engine types are Tika, Docling, Document Intelligence, Mistral OCR, Google Cloud Document AI, and LLM Vision — six in total (see Content Extraction for type descriptions). You can register multiple instances of the same type with different settings.
2
Use it in a profile
Pick the registered instance as the Default Engine of a document processing profile, or attach it to a specific extension under Extension → Engine Mapping.
An instance name is whatever you type when registering it. You may see names like Default - Primary or LLM Semantic - Primary on screen — that - Primary is part of a name an operator chose, not a label the system adds. Cloosphere has no feature for designating an instance as “Primary”: the engine for extensions that aren’t mapped is set per profile via Default Engine.
Document Processing Profiles
New feature — Bundle document extraction methods and chunking strategies into profiles. Create multiple profiles and select per Knowledge Base (KB).
What is a Profile?
Previously, only one global extraction/chunking setting was possible. With profiles, define different setting bundles per use case and pick the appropriate profile for each KB. Each profile consists of a default engine, an extraction engine mapping per file extension, and chunking settings.
In the Document Processing Profiles section at the top of the Documents settings page, create and manage profiles.
1
Create profile
Click + New Profile to open the profile creation modal.
Each profile sets a default engine, extension mapping, and chunking settings — there is no Primary step
Setting
Description
Default
Name
Profile identifier name
—
Default Engine
Engine used for extensions not listed in the mapping below. Pick Built-in Engine or one of the registered extraction engine instances
Built-in Engine
Extension → Engine Mapping
Use Add Extension to add a row and assign an engine per format group (PDF, Word, Excel, PowerPoint, Image, HTML, Text, Email, EPUB) or per custom extension you type in. A row can also select Built-in Engine
Chunks smaller than this are merged with the next one. Empty or 0 disables it
Empty
Max Tokens per Chunk
Chunks above this token count are re-split before embedding. Empty uses the global default; 0 disables it
Empty
Preserve Tables
Keep tables intact during chunking by attaching them to the preceding text chunk
On
Contextual Chunking
Add a context summary to each chunk to improve retrieval accuracy (increases processing cost)
See the warning below
Contextual chunking only takes effect when a Context Model is also selected. In a new profile the switch appears on, but with no model set it isn’t applied. Pick a model and save to use it. It also doesn’t apply to files attached ad hoc in chat — only to documents ingested into a Knowledge Base.
2
Set default profile
Click Set as Default in the profile list to sync that profile’s settings as the global default. KBs without a specified profile are processed with this default.
3
Pick profile in KB
In Workspace > Knowledge Base edit screen, pick the document processing profile. If unset, the default profile applies.
The default profile can’t be deleted. Set another profile as default before deleting.
Content Extraction
These are the six engine types you can register as Extraction Engine instances.
Register an instance, then pick it as a profile’s Default Engine or under Extension → Engine Mapping.
Engine Type
Strengths
Best For
Additional Settings
Tika
Apache Tika server-based
Diverse file format support
Server URL
Docling
Advanced document processing engine
Documents with complex layouts
Server URL
Document Intelligence
Azure AI Document Intelligence
Azure environments, accurate OCR
Endpoint + API Key
Google Cloud Document AI
Google Cloud Document AI
Tables, forms, multi-column layouts
Location + Processor ID + service account key (JSON). Project ID and Processor Version are optional
Mistral OCR
Mistral-based OCR
Image-based documents
Mistral API Key
LLM Vision
Vision LLM-based extraction
Complex layouts, text in charts/images
Vision model + extraction prompt (optional)
The Built-in Engine is not in this list. It isn’t something you register as an instance — it’s an option you pick directly in a profile’s Default Engine select and in mapping rows. If you register no extraction engines at all, every document is processed with this built-in engine.
The built-in engine is sufficient for most text-based documents (PDF, DOCX, TXT). For mostly scanned documents or image-heavy PDFs, Document Intelligence or Mistral OCR is recommended.
LLM Vision Extraction
New feature — A method that sends page images directly to a Vision-capable LLM (GPT-4o, Claude, etc.) for extraction as markdown text.
Compared to traditional OCR engines, this provides higher accuracy for complex layouts, text within images, and chart descriptions.Behavior:
Convert PDF to per-page images (rendered at 2× scale — roughly 144 DPI)
Normalize heading levels (##, ###) across the whole document, since pages were extracted independently
Setting
Description
Vision Model
Pick a Vision-supporting model registered in the system (e.g., gpt-4o)
Extraction Prompt
Custom prompt (default prompt used if empty)
LLM Vision incurs one LLM call per page. A 10-page PDF → ~11 calls (10 page extractions + 1 heading-level normalization). Consider processing time and cost for large documents.
Heading-level normalization sends only the heading lines, not the full body, and applies just the levels it gets back. Heading text itself is never changed, and if this call fails the extraction result is kept as-is. The step is skipped for single-page documents.
LLM Vision requires the pypdfium2 package. Without it, PDF uploads using this engine will fail. It handles PDFs and image files only — Office files such as PPTX and DOCX can’t be extracted with this engine.
Text Splitting
Configure how extracted text is split into search-friendly chunks.
Setting
Default
Description
Text Splitter
Character
Character: split by char count. Token: split by Tiktoken tokens. Semantic: meaning-based split
Chunk Size
1000
Chunk size (chars or tokens). Reference value in Semantic mode
Chunk Overlap
100
Overlap between chunks (preserves context flow)
How should I pick the Chunk Size?
Size
Pro
Con
Recommended
Small (≤ 500)
Precise search
Context may be cut
FAQ, short paragraphs
Medium (1000)
Balanced performance
—
Most cases (default)
Large (≥ 2000)
Wider context preserved
Lower search precision
Long narrative documents
Why is Chunk Overlap needed?
When sentences are cut at chunk boundaries, related content may be missed during search. Overlap shares some text between adjacent chunks to preserve context. The default 100 is appropriate in most cases.
Enabling Bypass Embedding and Retrieval skips text splitting and embedding, injecting the entire document directly into the LLM context. Use only for small documents — large documents may exceed token limits.
Semantic Chunking
New feature — A method that splits chunks based on inter-sentence meaning similarity instead of fixed size.
When Text Splitter is set to Semantic, sentences are converted to embedding vectors, and chunk boundaries are auto-determined where similarity between adjacent sentences drops sharply.
Fixed Size (Character/Token)
Semantic
Split criterion
Character / token count
Meaning similarity change
Pro
Fast and predictable
Splits at topic boundaries, improves search accuracy
Con
May cut mid-sentence/paragraph
Requires embedding engine, longer processing
Recommended
General documents
Long documents with frequent topic shifts
Semantic chunking requires an embedding engine to be configured. Without one, picking Semantic causes errors.
Table Preservation
New feature — Preserve tables (HTML/markdown format) in documents intact instead of splitting.
Standard chunking can split tables across multiple chunks, losing row/column information. With Table Preservation enabled:
Auto-detect tables in the document (HTML <table>, markdown |...|)
Standard chunking on text portions only
Tables are attached intact to the closest text chunk
Tables exceeding chunk size are split row-by-row while preserving the header
Controlled by the Preserve Tables option in profile advanced settings (default: enabled).
Especially effective for table-heavy financial reports and technical specifications. Search accuracy for specific cell values within tables is greatly improved.
Contextual Chunking
New feature — Each chunk has a context summary of the entire document generated by the LLM and prepended. Implements Anthropic’s Contextual Retrieval technique.
In later chunks of long documents, earlier context is lost, causing search matching failures. With contextual preservation enabled:
After chunking completes, call LLM for each chunk
Generate a summary of “where this chunk is in the entire document and its context”
Prepend the summary to the chunk before vectorization
[Before contextual preservation] Chunk: "In conclusion, the initial hypothesis was confirmed." → Can't tell what "A hypothesis" refers to → search fails[After contextual preservation] Chunk: "This chunk is the conclusion of the A Hypothesis Verification Report, containing the final judgment integrating 50 experiment results. In conclusion, the initial hypothesis was confirmed." → Includes context → search succeeds
Setting
Description
Contextual Chunking
Activate/deactivate toggle
Model
LLM model used for context summarization
LLM calls happen per chunk count. A document with 100 chunks → 100 LLM calls. Bulk uploading large documents can greatly increase API costs. Use selectively for small or important documents.
Embedding Settings
Configure the embedding engine and model that converts documents to vectors.
Setting
Default
Description
Embedding Model Engine
SentenceTransformers
Service for vector conversion
Embedding Model
all-MiniLM-L6-v2
Model name to use
Batch Size
32
Documents processed per batch (only shown for Ollama, OpenAI, Azure)
Embedding Dimension
0 (auto)
Vector dimension count. 0 uses model default
Changing the embedding model makes existing document vectors incompatible. After changing the model, all Knowledge Bases must be reindexed.
Supported Engines:
SentenceTransformers
OpenAI
Azure OpenAI
Ollama
Vertex AI
Gemini
Open-source embedding engine running locally.
Setting
Description
Model
HuggingFace model name (e.g., sentence-transformers/all-MiniLM-L6-v2)
Setting
Description
API URL
OpenAI API Endpoint
API Key
API authentication key
Model
e.g., text-embedding-3-small
Setting
Description
API URL
Azure OpenAI Endpoint
API Key
Azure auth key
API Version
API version
Setting
Description
API URL
Ollama server address
Model
e.g., nomic-embed-text
Setting
Description
Project ID
Google Cloud project ID
Location
Region (default: us-central1)
Service Account Key
Global key or Custom JSON key
Uses Google Gemini embedding API.
Setting
Description
API Key
Gemini API auth key
Model
Embedding model name (e.g., text-embedding-004)
File Upload Limits
Setting
Default
Description
Max file size
Unlimited
Single file upload max size (in MB)
Max files
Unlimited
Files usable simultaneously per chat
Allowed file extensions
All allowed
Comma-separated (e.g., pdf, docx, txt). Empty = all extensions allowed
PDF conversion extensions
Disabled
Extensions to convert to PDF via LibreOffice (e.g., pptx, docx, xlsx). LibreOffice installation required
Question Generation
A feature that pre-generates “questions a user might ask” for each chunk via LLM to improve search accuracy.
Environment variables must be set to enable toggles. Enabling a toggle shows the cloud source in the Knowledge Base’s Add Content menu.
Reindexing and Reset
The actions below are irreversible. Always confirm before executing.
Action
Description
Impact Scope
Reindex Knowledge Bases
Rebuild vector indexes for all KBs
All Knowledge Bases (takes time)
Reset Vector Storage
Delete all data in the Vector DB
Deletes all vectors → reindexing required
Reset Upload Directory
Delete all files uploaded to the server
Deletes original files
When Reindexing is Needed
When you’ve changed the embedding model
When you’ve changed the search engine (Vector DB)
When you’ve changed chunk size/overlap
When document processing has issues
Related Pages
Knowledge Base
Create Knowledge Bases and manage documents — pick profile per KB
Search Engine
Vector DB and search parameter settings
Dynamic Filters
Knowledge Base metadata filter settings
Was this page helpful?
{"docsConfig":{"theme":"mint","$schema":"https://mintlify.com/docs.json","name":"Cloosphere Guide","colors":{"primary":"#5f22d0","light":"#7c3aed","dark":"#4c1d95"},"logo":{"light":"https://mintcdn.com/cloocus/uXAlvLAzCJlwDYsB/images/logo.png?fit=max\u0026auto=format\u0026n=uXAlvLAzCJlwDYsB\u0026q=85\u0026s=93fb26230c58b4ba9411c3e97e20b439","dark":"https://mintcdn.com/cloocus/uXAlvLAzCJlwDYsB/images/logo.png?fit=max\u0026auto=format\u0026n=uXAlvLAzCJlwDYsB\u0026q=85\u0026s=93fb26230c58b4ba9411c3e97e20b439","href":"/ko/index"},"favicon":"https://mintcdn.com/cloocus/uXAlvLAzCJlwDYsB/images/favicon.png?fit=max\u0026auto=format\u0026n=uXAlvLAzCJlwDYsB\u0026q=85\u0026s=0e41ff0755aa623d52e26f279f285f71","appearance":{"default":"system","strict":false},"interaction":{"drilldown":true},"styling":{"latex":false},"navigation":{"languages":[{"language":"ko","default":true,"href":"ko"},{"language":"en","href":"en"}]}},"deployment":{"deploymentMetadataValue":{"subdomain":"cloocus","hasAudioNarration":false,"actualSubdomain":"cloocus","customDomains":["docs.cloosphere.ai"],"gitSource":{"type":"github","owner":"ais-cloocus","repo":"cloosphere-docs","deployBranch":"main","contentDirectory":"","isPrivate":true,"mountPath":"","label":""},"feedback":{"thumbs":true},"entitlements":{},"buildId":"6a954066df6b735599399868","clientVersion":"0.0.4270","searchFilterCounts":[{"language":"en","tag":"Admin","count":30},{"language":"ko","tag":"관리자","count":30},{"language":"en","tag":"Settings","count":24},{"language":"ko","tag":"설정","count":24},{"language":"en","tag":"Monitoring","count":21},{"language":"ko","tag":"모니터링","count":21},{"language":"en","tag":"Workspace","count":16},{"language":"ko","tag":"워크스페이스","count":16},{"language":"en","tag":"Chat","count":11},{"language":"ko","tag":"채팅","count":11},{"language":"en","tag":"Getting Started","count":10},{"language":"ko","tag":"시작하기","count":10},{"language":"en","tag":"Collaboration","count":8},{"language":"ko","tag":"협업","count":8},{"language":"en","tag":"Data \u0026 Operations","count":6},{"language":"ko","tag":"운영 · 지원","count":6},{"language":"ko","tag":"데이터 및 운영 보안","count":6},{"language":"en","tag":"AI Features \u0026 Safety","count":5},{"language":"en","tag":"UI \u0026 Channels","count":5},{"language":"ko","tag":"AI 기능 및 안전","count":5}]},"assistantConfig":{"enableAskAiButton":false,"enableStarterQuestions":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":[{"filePath":"styles.css","content":"/* latin-ext */\n@font-face {\n font-family: 'DM Sans';\n font-style: normal;\n font-weight: 400;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/dmsans/v17/rP2Yp2ywxg089UriI5-g4vlH9VoD8Cmcqbu6-K6h9Q.woff2) format('woff2');\n unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C4, U+2113, U+2C60-2C7F, U+A720-A7FF;\n}\n/* latin */\n@font-face {\n font-family: 'DM Sans';\n font-style: normal;\n font-weight: 400;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/dmsans/v17/rP2Yp2ywxg089UriI5-g4vlH9VoD8Cmcqbu0-K4.woff2) format('woff2');\n unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD;\n}\n/* latin-ext */\n@font-face {\n font-family: 'DM Sans';\n font-style: normal;\n font-weight: 500;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/dmsans/v17/rP2Yp2ywxg089UriI5-g4vlH9VoD8Cmcqbu6-K6h9Q.woff2) format('woff2');\n unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C4, U+2113, U+2C60-2C7F, U+A720-A7FF;\n}\n/* latin */\n@font-face {\n font-family: 'DM Sans';\n font-style: normal;\n font-weight: 500;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/dmsans/v17/rP2Yp2ywxg089UriI5-g4vlH9VoD8Cmcqbu0-K4.woff2) format('woff2');\n unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD;\n}\n/* latin-ext */\n@font-face {\n font-family: 'DM Sans';\n font-style: normal;\n font-weight: 600;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/dmsans/v17/rP2Yp2ywxg089UriI5-g4vlH9VoD8Cmcqbu6-K6h9Q.woff2) format('woff2');\n unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C4, U+2113, U+2C60-2C7F, U+A720-A7FF;\n}\n/* latin */\n@font-face {\n font-family: 'DM Sans';\n font-style: normal;\n font-weight: 600;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/dmsans/v17/rP2Yp2ywxg089UriI5-g4vlH9VoD8Cmcqbu0-K4.woff2) format('woff2');\n unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD;\n}\n/* vietnamese */\n@font-face {\n font-family: 'Space Grotesk';\n font-style: normal;\n font-weight: 500;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/spacegrotesk/v22/V8mDoQDjQSkFtoMM3T6r8E7mPb54C-s0.woff2) format('woff2');\n unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB;\n}\n/* latin-ext */\n@font-face {\n font-family: 'Space Grotesk';\n font-style: normal;\n font-weight: 500;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/spacegrotesk/v22/V8mDoQDjQSkFtoMM3T6r8E7mPb94C-s0.woff2) format('woff2');\n unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C4, U+2113, U+2C60-2C7F, U+A720-A7FF;\n}\n/* latin */\n@font-face {\n font-family: 'Space Grotesk';\n font-style: normal;\n font-weight: 500;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/spacegrotesk/v22/V8mDoQDjQSkFtoMM3T6r8E7mPbF4Cw.woff2) format('woff2');\n unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD;\n}\n/* vietnamese */\n@font-face {\n font-family: 'Space Grotesk';\n font-style: normal;\n font-weight: 600;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/spacegrotesk/v22/V8mDoQDjQSkFtoMM3T6r8E7mPb54C-s0.woff2) format('woff2');\n unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB;\n}\n/* latin-ext */\n@font-face {\n font-family: 'Space Grotesk';\n font-style: normal;\n font-weight: 600;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/spacegrotesk/v22/V8mDoQDjQSkFtoMM3T6r8E7mPb94C-s0.woff2) format('woff2');\n unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C4, U+2113, U+2C60-2C7F, U+A720-A7FF;\n}\n/* latin */\n@font-face {\n font-family: 'Space Grotesk';\n font-style: normal;\n font-weight: 600;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/spacegrotesk/v22/V8mDoQDjQSkFtoMM3T6r8E7mPbF4Cw.woff2) format('woff2');\n unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD;\n}\n/* vietnamese */\n@font-face {\n font-family: 'Space Grotesk';\n font-style: normal;\n font-weight: 700;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/spacegrotesk/v22/V8mDoQDjQSkFtoMM3T6r8E7mPb54C-s0.woff2) format('woff2');\n unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB;\n}\n/* latin-ext */\n@font-face {\n font-family: 'Space Grotesk';\n font-style: normal;\n font-weight: 700;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/spacegrotesk/v22/V8mDoQDjQSkFtoMM3T6r8E7mPb94C-s0.woff2) format('woff2');\n unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C4, U+2113, U+2C60-2C7F, U+A720-A7FF;\n}\n/* latin */\n@font-face {\n font-family: 'Space Grotesk';\n font-style: normal;\n font-weight: 700;\n font-display: swap;\n src: url(/_mintlify/assets/fonts/google/s/spacegrotesk/v22/V8mDoQDjQSkFtoMM3T6r8E7mPbF4Cw.woff2) format('woff2');\n unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD;\n}\n:root{--brand-primary: #5f22d0;--brand-light: #7c3aed;--brand-dark: #4c1d95;--brand-subtle: #ede9fe;--brand-glow: rgba(95, 34, 208, 0.15);--page-text: #0f172a;--page-muted: #6b7280}body{font-family:\"DM Sans\", sans-serif;color:var(--page-text);font-size:17px}.prose h1,.prose h2,.prose h3,.prose h4{font-family:\"Space Grotesk\", \"DM Sans\", sans-serif;letter-spacing:-0.015em}.prose h1{font-size:44px}.prose h2{font-size:28px;margin-top:36px;margin-bottom:10px}.prose h3{font-size:22px;margin-top:28px;margin-bottom:8px}.prose p,.prose li{color:var(--page-text);line-height:1.75;font-size:1rem}.prose hr{border:0;height:1px;background:linear-gradient(90deg, rgba(148, 163, 184, 0.4), rgba(148, 163, 184, 0.05));margin:28px 0}.dark body{color:#e2e8f0}.dark .prose p,.dark .prose li{color:#e2e8f0}.dark .prose hr{background:linear-gradient(90deg, rgba(148, 163, 184, 0.3), rgba(148, 163, 184, 0.05))}[class*=\"Card\"] svg,.card svg,a[href] svg{color:var(--brand-primary)!important}.card [data-component-part=\"card-icon\"]{margin-bottom:8px}.card [data-component-part=\"card-title\"]{margin-top:12px!important}[class*=\"Card\"],.card{border-radius:16px!important;border-color:rgba(148, 163, 184, 0.25)!important;transition:border-color 0.2s ease, box-shadow 0.2s ease, transform 0.2s ease}[class*=\"Card\"]:hover,.card:hover{transform:translateY(-2px);border-color:var(--brand-light)!important;box-shadow:0 4px 20px var(--brand-glow)!important}a:not([class*=\"Card\"]):hover{color:var(--brand-primary)!important}.stable-scrollbar-gutter.overflow-auto{width:220px!important;min-width:220px!important}.stable-scrollbar-gutter.overflow-auto+div,.stable-scrollbar-gutter.overflow-auto~main{flex:1!important}.flex.flex-row-reverse.gap-12{gap:32px!important}.xl\\:flex.self-start.sticky{max-width:15rem!important}.xl\\:flex.box-border.max-h-full.pl-10{width:14rem!important;padding-left:24px!important}.text-sm.leading-6.w-\\[16\\.5rem\\]{width:100%!important;font-size:13px!important;line-height:1.6!important}#table-of-contents-content a{padding:5px 0!important;display:block}#table-of-contents-content,#table-of-contents-content ul{list-style:none!important;padding-left:0!important;margin-left:0!important}#table-of-contents-content li::marker{content:\"\"!important}#table-of-contents-content>li{border-left:2px solid rgba(148, 163, 184, 0.1);padding-left:12px;transition:border-color 0.15s ease}.dark #table-of-contents-content>li{border-left-color:rgba(148, 163, 184, 0.08)}#table-of-contents-content a{color:var(--page-muted)!important;font-weight:400!important;transition:color 0.15s ease}.dark #table-of-contents-content a{color:#94a3b8!important}#table-of-contents-content>li>a{color:var(--page-text)!important;font-weight:500!important}.dark #table-of-contents-content>li>a{color:#e2e8f0!important}#table-of-contents-content>li>ul{padding-left:12px!important}#table-of-contents-content a:hover{color:var(--brand-dark)!important}.dark #table-of-contents-content a:hover{color:#e2e8f0!important}#table-of-contents-content a.text-primary,#table-of-contents-content a.font-semibold,#table-of-contents-content>li.is-active-item>a{color:var(--brand-primary)!important;font-weight:600!important}.dark #table-of-contents-content a.text-primary,.dark #table-of-contents-content a.font-semibold,.dark #table-of-contents-content a[class*=\"dark:text-primary\"],.dark #table-of-contents-content>li.is-active-item>a{color:var(--brand-light)!important;font-weight:600!important}#table-of-contents-content>li.is-active-item{border-left-color:var(--brand-primary)!important}.dark #table-of-contents-content>li.is-active-item{border-left-color:var(--brand-light)!important}#table-of-contents-content>li:has(>a.border-primary),#table-of-contents-content>li:has(>a.text-primary){border-left-color:var(--brand-primary)!important}.dark #table-of-contents-content>li:has(>a.border-primary),.dark #table-of-contents-content>li:has(>a.text-primary),.dark #table-of-contents-content>li:has(>a[class*=\"dark:text-primary\"]),.dark #table-of-contents-content>li:has(>a[class*=\"dark:border-primary\"]){border-left-color:var(--brand-light)!important}.landing-hero{text-align:center;padding:72px 32px 24px;max-width:860px;margin:0 auto}.landing-hero h1{font-family:\"Space Grotesk\", \"DM Sans\", sans-serif;font-size:52px;font-weight:700;letter-spacing:-0.03em;margin:0 0 20px;background:linear-gradient(135deg, var(--brand-dark), var(--brand-primary), var(--brand-light));-webkit-background-clip:text;-webkit-text-fill-color:transparent;background-clip:text}.dark .landing-hero h1{background:linear-gradient(135deg, var(--brand-light), #a78bfa, #c4b5fd);-webkit-background-clip:text;background-clip:text}.landing-hero p{font-size:19px;line-height:1.7;color:var(--page-muted);margin:0}.dark .landing-hero p{color:#94a3b8}.landing-section{max-width:1280px;margin:0 auto;padding:0 40px 56px}.landing-section h2{font-family:\"Space Grotesk\", \"DM Sans\", sans-serif;font-size:26px;font-weight:600;letter-spacing:-0.015em;margin:0 0 6px;color:var(--page-text)}.dark .landing-section h2{color:#e2e8f0}.landing-section>p{color:var(--page-muted);font-size:16px;margin:0 0 24px}.dark .landing-section>p{color:#94a3b8}.landing-cta{max-width:1280px;margin:0 auto;padding:0 40px 48px}.landing-divider{max-width:1280px;margin:0 auto 48px;padding:0 40px}.landing-divider hr{border:0;height:1px;background:linear-gradient(90deg, transparent, rgba(148, 163, 184, 0.3), transparent);margin:0}.dark .landing-divider hr{background:linear-gradient(90deg, transparent, rgba(148, 163, 184, 0.15), transparent)}.landing-section [role=\"tablist\"]{margin-bottom:12px}.landing-section [role=\"tabpanel\"]>p:first-child{color:var(--page-muted);font-size:15px;margin-bottom:20px}.dark .landing-section [role=\"tabpanel\"]>p:first-child{color:#94a3b8}.group:has(>.mermaid){background:linear-gradient(135deg, #faf8ff 0%, #f3f0ff 100%);border:1px solid rgba(95, 34, 208, 0.08);border-radius:12px;padding:28px 24px;margin:24px 0}.dark .group:has(>.mermaid){background:linear-gradient(135deg, rgba(95, 34, 208, 0.06) 0%, rgba(95, 34, 208, 0.03) 100%);border-color:rgba(124, 58, 237, 0.12)}.mermaid svg .node rect,.mermaid svg .node circle,.mermaid svg .node ellipse,.mermaid svg .node polygon,.mermaid svg .node path{fill:#ede9fe!important;stroke:#7c3aed!important;stroke-width:1.5px!important;rx:8px}.dark .mermaid svg .node rect,.dark .mermaid svg .node circle,.dark .mermaid svg .node ellipse,.dark .mermaid svg .node polygon,.dark .mermaid svg .node path{fill:rgba(124, 58, 237, 0.15)!important;stroke:#a78bfa!important}.mermaid svg .node.rhombus rect,.mermaid svg .node polygon[points]{fill:#ddd6fe!important;stroke:#6d28d9!important}.dark .mermaid svg .node.rhombus rect,.dark .mermaid svg .node polygon[points]{fill:rgba(109, 40, 217, 0.2)!important;stroke:#a78bfa!important}.mermaid svg .cluster rect{fill:rgba(237, 233, 254, 0.4)!important;stroke:rgba(124, 58, 237, 0.25)!important;stroke-width:1.5px!important;rx:10px!important}.dark .mermaid svg .cluster rect{fill:rgba(124, 58, 237, 0.08)!important;stroke:rgba(167, 139, 250, 0.2)!important}.mermaid svg .cluster-label .nodeLabel,.mermaid svg .cluster span{fill:var(--brand-dark)!important;color:var(--brand-dark)!important;font-weight:600!important}.dark .mermaid svg .cluster-label .nodeLabel,.dark .mermaid svg .cluster span{fill:#c4b5fd!important;color:#c4b5fd!important}.mermaid svg .edgePath path.path{stroke:#8b5cf6!important;stroke-width:1.5px!important}.dark .mermaid svg .edgePath path.path{stroke:#a78bfa!important}.mermaid svg .marker{fill:#8b5cf6!important;stroke:#8b5cf6!important}.dark .mermaid svg .marker{fill:#a78bfa!important;stroke:#a78bfa!important}.mermaid svg .label,.mermaid svg .nodeLabel,.mermaid svg .label text,.mermaid svg span{fill:#1e1b4b!important;color:#1e1b4b!important;font-family:\"DM Sans\", sans-serif!important}.dark .mermaid svg .label,.dark .mermaid svg .nodeLabel,.dark .mermaid svg .label text,.dark .mermaid svg span{fill:#e2e8f0!important;color:#e2e8f0!important}.mermaid svg .edgeLabel{background-color:#faf8ff!important}.dark .mermaid svg .edgeLabel{background-color:transparent!important}.mermaid svg .edgeLabel span{fill:#6d28d9!important;color:#6d28d9!important;font-size:13px!important}.dark .mermaid svg .edgeLabel span{fill:#c4b5fd!important;color:#c4b5fd!important}.mermaid svg{background:transparent!important}#localization-select-trigger>div:first-child{display:none!important}[data-relocated-lang]{gap:0!important}::selection{background:var(--brand-subtle);color:var(--brand-dark)}.dark ::selection{background:var(--brand-dark);color:#ffffff}*:focus-visible{outline-color:var(--brand-primary)!important}::-webkit-scrollbar-thumb{background:var(--brand-light);border-radius:4px}::-webkit-scrollbar-thumb:hover{background:var(--brand-primary)}.cloo-badge{display:inline-block;font-family:\"DM Sans\", sans-serif;font-size:12px;font-weight:600;line-height:1.5;padding:0 8px;border-radius:6px;letter-spacing:-0.01em;vertical-align:middle;white-space:nowrap;margin-right:2px}.cloo-badge--required{background:#fef2f2;color:#b91c1c}.cloo-badge--multiworker{background:#fff7ed;color:#c2410c}.cloo-badge--recommended{background:#eff6ff;color:#1d4ed8}.cloo-badge--optional{background:#f1f5f9;color:#475569}.dark .cloo-badge--required{background:rgba(185, 28, 28, 0.2);color:#fca5a5}.dark .cloo-badge--multiworker{background:rgba(194, 65, 12, 0.2);color:#fdba74}.dark .cloo-badge--recommended{background:rgba(29, 78, 216, 0.22);color:#93c5fd}.dark .cloo-badge--optional{background:rgba(71, 85, 105, 0.28);color:#cbd5e1}"}]}
{"page":{"pagePath":"en/admin/settings/documents","slug":"/en/admin/settings/documents","pageMetadata":{"title":"Documents","description":"Configure document processing profiles, content extraction, embedding engines, and search parameters.","href":"/en/admin/settings/documents"},"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: \"Configure document processing profiles, content extraction, embedding engines, and search parameters.\"\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`,li:`li`,ol:`ol`,p:`p`,pre:`pre`,span:`span`,strong:`strong`,tbody:`tbody`,td:`td`,th:`th`,thead:`thead`,tr:`tr`,ul:`ul`,..._provideComponents(),...props.components},{Accordion,AccordionGroup,Card,CodeBlock,Columns,Danger,Frame,Heading,Info,Note,OptimizedImage,Step,Steps,Tab,Table,Tabs,Tip,Warning}=_components;if(!Accordion)_missingMdxReference(`Accordion`,true);if(!AccordionGroup)_missingMdxReference(`AccordionGroup`,true);if(!Card)_missingMdxReference(`Card`,true);if(!CodeBlock)_missingMdxReference(`CodeBlock`,true);if(!Columns)_missingMdxReference(`Columns`,true);if(!Danger)_missingMdxReference(`Danger`,true);if(!Frame)_missingMdxReference(`Frame`,true);if(!Heading)_missingMdxReference(`Heading`,true);if(!Info)_missingMdxReference(`Info`,true);if(!Note)_missingMdxReference(`Note`,true);if(!OptimizedImage)_missingMdxReference(`OptimizedImage`,true);if(!Step)_missingMdxReference(`Step`,true);if(!Steps)_missingMdxReference(`Steps`,true);if(!Tab)_missingMdxReference(`Tab`,true);if(!Table)_missingMdxReference(`Table`,true);if(!Tabs)_missingMdxReference(`Tabs`,true);if(!Tip)_missingMdxReference(`Tip`,true);if(!Warning)_missingMdxReference(`Warning`,true);return _jsxs(_Fragment,{children:[_jsx(Info,{children:`Admin › Settings › Documents`}),`\n`,_jsx(_components.p,{children:`The document processing quality of Knowledge Bases (KBSphere) is determined by these settings.`}),`\n`,_jsxs(_components.ul,{children:[`\n`,_jsx(_components.li,{children:`Properly configuring content extraction engines, text splitting, and embedding models is essential for RAG search accuracy.`}),`\n`]}),`\n`,_jsx(Frame,{caption:`Manage all RAG pipeline settings in Admin > Settings > Documents`,children:_jsx(OptimizedImage,{src:`https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-full.png?fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=b04567fbb785fb5786bcb17757c8009d`,alt:`Documents settings tab`,\"data-og-width\":`2880`,width:`2880`,\"data-og-height\":`1800`,height:`1800`,\"data-path\":`images/en/admin/settings-documents-full.png`,\"data-optimize\":`true`,\"data-opv\":`3`,srcset:`https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-full.png?w=280\u0026fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=256030b9191924ed7659f11d9de70d21 280w, https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-full.png?w=560\u0026fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=849d67c3e0c9d70e29cc077e04cd332e 560w, https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-full.png?w=840\u0026fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=4c5248abc807b206b03eb1c7fac1f7a6 840w, https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-full.png?w=1100\u0026fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=d68cdcbbc6fba555517d9a82ec738748 1100w, https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-full.png?w=1650\u0026fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=8d99c50f5bd831fdd0d5520755d5da5d 1650w, https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-full.png?w=2500\u0026fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=946d725f2608ce158ca281fb262b1cfb 2500w`})}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`extraction-engine`,children:`Extraction Engine`}),`\n`,_jsxs(_components.p,{children:[`In the `,_jsx(_components.strong,{children:`Extraction Engine`}),` section at the top of the page, register and manage the `,_jsx(_components.strong,{children:`engine instances`}),` used to extract document text.`]}),`\n`,_jsxs(_components.ul,{children:[`\n`,_jsxs(_components.li,{children:[`Extraction is not “picking a single engine globally” — instead, you register `,_jsx(_components.strong,{children:`named instances`}),` and select them per extension within a profile.`]}),`\n`]}),`\n`,_jsxs(Steps,{children:[_jsx(Step,{title:`Register an engine instance`,children:_jsxs(_components.p,{children:[`Add an instance with the `,_jsx(_components.strong,{children:`+`}),` button in the `,_jsx(_components.strong,{children:`Extraction Engine`}),` section, then specify a `,_jsx(_components.strong,{children:`Name`}),` and an `,_jsx(_components.strong,{children:`Engine Type`}),`. The selectable engine types are `,_jsx(_components.code,{children:`Tika`}),`, `,_jsx(_components.code,{children:`Docling`}),`, `,_jsx(_components.code,{children:`Document Intelligence`}),`, `,_jsx(_components.code,{children:`Mistral OCR`}),`, `,_jsx(_components.code,{children:`Google Cloud Document AI`}),`, and `,_jsx(_components.code,{children:`LLM Vision`}),` — six in total (see `,_jsx(_components.a,{href:`#content-extraction`,children:`Content Extraction`}),` for type descriptions). You can register multiple instances of the same type with different settings.`]})}),_jsx(Step,{title:`Use it in a profile`,children:_jsxs(_components.p,{children:[`Pick the registered instance as the `,_jsx(_components.strong,{children:`Default Engine`}),` of a `,_jsx(_components.a,{href:`#document-processing-profiles`,children:`document processing profile`}),`, or attach it to a specific extension under `,_jsx(_components.strong,{children:`Extension → Engine Mapping`}),`.`]})})]}),`\n`,_jsx(Note,{children:_jsxs(_components.p,{children:[_jsx(_components.strong,{children:`An instance name is whatever you type when registering it.`}),` You may see names like `,_jsx(_components.code,{children:`Default - Primary`}),` or `,_jsx(_components.code,{children:`LLM Semantic - Primary`}),` on screen — that `,_jsx(_components.code,{children:`- Primary`}),` is part of a name an operator chose, not a label the system adds. Cloosphere has no feature for designating an instance as “Primary”: the engine for extensions that aren’t mapped is set `,_jsx(_components.strong,{children:`per profile`}),` via `,_jsx(_components.code,{children:`Default Engine`}),`.`]})}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`document-processing-profiles`,children:`Document Processing Profiles`}),`\n`,_jsx(Info,{children:_jsxs(_components.p,{children:[_jsx(_components.strong,{children:`New feature`}),` — Bundle document extraction methods and chunking strategies into `,_jsx(_components.strong,{children:`profiles`}),`. Create multiple profiles and select per Knowledge Base (KB).`]})}),`\n`,_jsx(Heading,{level:`3`,id:`what-is-a-profile`,children:`What is a Profile?`}),`\n`,_jsxs(_components.p,{children:[`Previously, only one global extraction/chunking setting was possible. With profiles, define `,_jsx(_components.strong,{children:`different setting bundles per use case`}),` and pick the appropriate profile for each KB. Each profile consists of a `,_jsx(_components.strong,{children:`default engine`}),`, an `,_jsx(_components.strong,{children:`extraction engine mapping per file extension`}),`, and `,_jsx(_components.strong,{children:`chunking settings`}),`.`]}),`\n`,_jsx(CodeBlock,{numberOfLines:`5`,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:`5`,children:[_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:`Examples:`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:` • \"Default Extraction\" → built-in engine + fixed-size chunking`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:` • \"High-Precision OCR\" → Document Intelligence + table preservation`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:` • \"LLM Vision + Context\" → Vision LLM extraction + semantic chunking + Contextual Chunking`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{})})]})})}),`\n`,_jsx(Heading,{level:`3`,id:`profile-management`,children:`Profile Management`}),`\n`,_jsxs(_components.p,{children:[`In the `,_jsx(_components.strong,{children:`Document Processing Profiles`}),` section at the top of the Documents settings page, create and manage profiles.`]}),`\n`,_jsxs(Steps,{children:[_jsxs(Step,{title:`Create profile`,children:[_jsxs(_components.p,{children:[`Click `,_jsx(_components.strong,{children:`+ New Profile`}),` to open the profile creation modal.`]}),_jsx(Frame,{caption:`Each profile sets a default engine, extension mapping, and chunking settings — there is no Primary step`,children:_jsx(OptimizedImage,{src:`https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-profile-modal.png?fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=83e0819187c41c687bcdc4a500d0ee48`,alt:`Add document processing profile modal`,\"data-og-width\":`2880`,width:`2880`,\"data-og-height\":`1800`,height:`1800`,\"data-path\":`images/en/admin/settings-documents-profile-modal.png`,\"data-optimize\":`true`,\"data-opv\":`3`,srcset:`https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-profile-modal.png?w=280\u0026fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=724dca3d36dbed711b16c5357a191dca 280w, https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-profile-modal.png?w=560\u0026fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=33cc79c99a78369f3f4e952893b1a48e 560w, https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-profile-modal.png?w=840\u0026fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=2e95b4323f8253e09b02347d3734ba7f 840w, https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-profile-modal.png?w=1100\u0026fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=5cf322af9de5bdbd39c3c09c9c5236f0 1100w, https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-profile-modal.png?w=1650\u0026fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=51aa45a457434dc430ffddf85cc1146e 1650w, https://mintcdn.com/cloocus/V0v-rkd8mEQoP_Pd/images/en/admin/settings-documents-profile-modal.png?w=2500\u0026fit=max\u0026auto=format\u0026n=V0v-rkd8mEQoP_Pd\u0026q=85\u0026s=1c40a5a298a59b720bceec8db9f4cc7b 2500w`})}),_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Setting`}),_jsx(_components.th,{children:`Description`}),_jsx(_components.th,{style:{textAlign:`center`},children:`Default`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Name`})}),_jsx(_components.td,{children:`Profile identifier name`}),_jsx(_components.td,{style:{textAlign:`center`},children:`—`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Default Engine`})}),_jsxs(_components.td,{children:[`Engine used for extensions not listed in the mapping below. Pick `,_jsx(_components.code,{children:`Built-in Engine`}),` or one of the registered `,_jsx(_components.a,{href:`#extraction-engine`,children:`extraction engine instances`})]}),_jsx(_components.td,{style:{textAlign:`center`},children:`Built-in Engine`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Extension → Engine Mapping`})}),_jsxs(_components.td,{children:[`Use `,_jsx(_components.code,{children:`Add Extension`}),` to add a row and assign an engine per `,_jsx(_components.strong,{children:`format group`}),` (PDF, Word, Excel, PowerPoint, Image, HTML, Text, Email, EPUB) or per custom extension you type in. A row can also select `,_jsx(_components.code,{children:`Built-in Engine`})]}),_jsx(_components.td,{style:{textAlign:`center`},children:`Empty`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Text Splitter`})}),_jsxs(_components.td,{children:[`Chunking strategy (`,_jsx(_components.code,{children:`Default (Character)`}),` / `,_jsx(_components.code,{children:`Token (Tiktoken)`}),` / `,_jsx(_components.code,{children:`Semantic`}),`)`]}),_jsx(_components.td,{style:{textAlign:`center`},children:`Default (Character)`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Chunk Size`})}),_jsx(_components.td,{children:`Size of each chunk`}),_jsx(_components.td,{\"data-numeric\":`true`,style:{textAlign:`center`},children:`1000`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Chunk Overlap`})}),_jsx(_components.td,{children:`Overlap between chunks`}),_jsx(_components.td,{\"data-numeric\":`true`,style:{textAlign:`center`},children:`100`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Min Tokens per Chunk`})}),_jsxs(_components.td,{children:[`Chunks smaller than this are merged with the next one. Empty or `,_jsx(_components.code,{children:`0`}),` disables it`]}),_jsx(_components.td,{style:{textAlign:`center`},children:`Empty`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Max Tokens per Chunk`})}),_jsxs(_components.td,{children:[`Chunks above this token count are re-split before embedding. Empty uses the global default; `,_jsx(_components.code,{children:`0`}),` disables it`]}),_jsx(_components.td,{style:{textAlign:`center`},children:`Empty`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Preserve Tables`})}),_jsx(_components.td,{children:`Keep tables intact during chunking by attaching them to the preceding text chunk`}),_jsx(_components.td,{style:{textAlign:`center`},children:`On`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Contextual Chunking`})}),_jsx(_components.td,{children:`Add a context summary to each chunk to improve retrieval accuracy (increases processing cost)`}),_jsx(_components.td,{style:{textAlign:`center`},children:`See the warning below`})]})]})]}),_jsx(Warning,{children:_jsxs(_components.p,{children:[_jsxs(_components.strong,{children:[`Contextual chunking only takes effect when a `,_jsx(_components.code,{children:`Context Model`}),` is also selected.`]}),` In a new profile the switch appears on, but with no model set it isn’t applied. Pick a model and save to use it. It also doesn’t apply to files attached ad hoc in chat — only to documents ingested into a Knowledge Base.`]})})]}),_jsx(Step,{title:`Set default profile`,children:_jsxs(_components.p,{children:[`Click `,_jsx(_components.strong,{children:`Set as Default`}),` in the profile list to sync that profile’s settings as the `,_jsx(_components.strong,{children:`global default`}),`. KBs without a specified profile are processed with this default.`]})}),_jsx(Step,{title:`Pick profile in KB`,children:_jsxs(_components.p,{children:[`In `,_jsx(_components.strong,{children:`Workspace > Knowledge Base`}),` edit screen, pick the document processing profile. If unset, the default profile applies.`]})})]}),`\n`,_jsx(Warning,{children:_jsx(_components.p,{children:`The default profile can’t be deleted. Set another profile as default before deleting.`})}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`content-extraction`,children:`Content Extraction`}),`\n`,_jsxs(_components.p,{children:[`These are the six `,_jsx(_components.strong,{children:`engine types`}),` you can register as `,_jsx(_components.a,{href:`#extraction-engine`,children:`Extraction Engine`}),` instances.`]}),`\n`,_jsxs(_components.ul,{children:[`\n`,_jsxs(_components.li,{children:[`Register an instance, then pick it as a profile’s `,_jsx(_components.code,{children:`Default Engine`}),` or under `,_jsx(_components.code,{children:`Extension → Engine Mapping`}),`.`]}),`\n`]}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Engine Type`}),_jsx(_components.th,{children:`Strengths`}),_jsx(_components.th,{children:`Best For`}),_jsx(_components.th,{children:`Additional Settings`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Tika`})}),_jsx(_components.td,{children:`Apache Tika server-based`}),_jsx(_components.td,{children:`Diverse file format support`}),_jsx(_components.td,{children:`Server URL`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Docling`})}),_jsx(_components.td,{children:`Advanced document processing engine`}),_jsx(_components.td,{children:`Documents with complex layouts`}),_jsx(_components.td,{children:`Server URL`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Document Intelligence`})}),_jsx(_components.td,{children:`Azure AI Document Intelligence`}),_jsx(_components.td,{children:`Azure environments, accurate OCR`}),_jsx(_components.td,{children:`Endpoint + API Key`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Google Cloud Document AI`})}),_jsx(_components.td,{children:`Google Cloud Document AI`}),_jsx(_components.td,{children:`Tables, forms, multi-column layouts`}),_jsx(_components.td,{children:`Location + Processor ID + service account key (JSON). Project ID and Processor Version are optional`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Mistral OCR`})}),_jsx(_components.td,{children:`Mistral-based OCR`}),_jsx(_components.td,{children:`Image-based documents`}),_jsx(_components.td,{children:`Mistral API Key`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`LLM Vision`})}),_jsx(_components.td,{children:`Vision LLM-based extraction`}),_jsx(_components.td,{children:`Complex layouts, text in charts/images`}),_jsx(_components.td,{children:`Vision model + extraction prompt (optional)`})]})]})]}),`\n`,_jsx(Note,{children:_jsxs(_components.p,{children:[`The `,_jsx(_components.strong,{children:`Built-in Engine`}),` is not in this list. It isn’t something you register as an instance — it’s an option you pick directly in a profile’s `,_jsx(_components.code,{children:`Default Engine`}),` select and in mapping rows. If you register no extraction engines at all, every document is processed with this built-in engine.`]})}),`\n`,_jsx(Tip,{children:_jsxs(_components.p,{children:[`The built-in engine is sufficient for most text-based documents (PDF, DOCX, TXT). For mostly scanned documents or image-heavy PDFs, `,_jsx(_components.strong,{children:`Document Intelligence`}),` or `,_jsx(_components.strong,{children:`Mistral OCR`}),` is recommended.`]})}),`\n`,_jsx(Heading,{level:`3`,id:`llm-vision-extraction`,children:`LLM Vision Extraction`}),`\n`,_jsx(Info,{children:_jsxs(_components.p,{children:[_jsx(_components.strong,{children:`New feature`}),` — A method that sends page images directly to a Vision-capable LLM (GPT-4o, Claude, etc.) for extraction as markdown text.`]})}),`\n`,_jsxs(_components.p,{children:[`Compared to traditional OCR engines, this provides higher accuracy for `,_jsx(_components.strong,{children:`complex layouts, text within images, and chart descriptions`}),`.`]}),`\n`,_jsx(_components.p,{children:_jsx(_components.strong,{children:`Behavior:`})}),`\n`,_jsxs(_components.ol,{children:[`\n`,_jsx(_components.li,{children:`Convert PDF to per-page images (rendered at 2× scale — roughly 144 DPI)`}),`\n`,_jsx(_components.li,{children:`Send each page image to Vision LLM in parallel`}),`\n`,_jsx(_components.li,{children:`LLM returns markdown preserving table/list/title structure`}),`\n`,_jsxs(_components.li,{children:[_jsxs(_components.strong,{children:[`Normalize heading levels (`,_jsx(_components.code,{children:`##`}),`, `,_jsx(_components.code,{children:`###`}),`)`]}),` across the whole document, since pages were extracted independently`]}),`\n`]}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Setting`}),_jsx(_components.th,{children:`Description`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Vision Model`})}),_jsxs(_components.td,{children:[`Pick a Vision-supporting model registered in the system (e.g., `,_jsx(_components.code,{children:`gpt-4o`}),`)`]})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Extraction Prompt`})}),_jsx(_components.td,{children:`Custom prompt (default prompt used if empty)`})]})]})]}),`\n`,_jsx(Warning,{children:_jsxs(_components.p,{children:[`LLM Vision incurs `,_jsx(_components.strong,{children:`one LLM call per page`}),`. A 10-page PDF → ~11 calls (10 page extractions + 1 heading-level normalization). Consider processing time and cost for large documents.`]})}),`\n`,_jsx(Note,{children:_jsxs(_components.p,{children:[`Heading-level normalization sends only the `,_jsx(_components.strong,{children:`heading lines`}),`, not the full body, and applies just the levels it gets back. Heading text itself is never changed, and if this call fails the extraction result is kept as-is. The step is skipped for single-page documents.`]})}),`\n`,_jsx(Note,{children:_jsxs(_components.p,{children:[`LLM Vision requires the `,_jsx(_components.strong,{children:_jsx(_components.code,{children:`pypdfium2`})}),` package. Without it, PDF uploads using this engine will fail. It handles `,_jsx(_components.strong,{children:`PDFs and image files`}),` only — Office files such as PPTX and DOCX can’t be extracted with this engine.`]})}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`text-splitting`,children:`Text Splitting`}),`\n`,_jsx(_components.p,{children:`Configure how extracted text is split into search-friendly chunks.`}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Setting`}),_jsx(_components.th,{style:{textAlign:`center`},children:`Default`}),_jsx(_components.th,{children:`Description`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Text Splitter`})}),_jsx(_components.td,{style:{textAlign:`center`},children:`Character`}),_jsxs(_components.td,{children:[_jsx(_components.code,{children:`Character`}),`: split by char count. `,_jsx(_components.code,{children:`Token`}),`: split by Tiktoken tokens. `,_jsx(_components.code,{children:`Semantic`}),`: meaning-based split`]})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Chunk Size`})}),_jsx(_components.td,{\"data-numeric\":`true`,style:{textAlign:`center`},children:`1000`}),_jsx(_components.td,{children:`Chunk size (chars or tokens). Reference value in Semantic mode`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Chunk Overlap`})}),_jsx(_components.td,{\"data-numeric\":`true`,style:{textAlign:`center`},children:`100`}),_jsx(_components.td,{children:`Overlap between chunks (preserves context flow)`})]})]})]}),`\n`,_jsxs(AccordionGroup,{children:[_jsx(Accordion,{title:`How should I pick the Chunk Size?`,icon:`circle-question`,id:`how-should-i-pick-the-chunk-size`,children:_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{style:{textAlign:`center`},children:`Size`}),_jsx(_components.th,{children:`Pro`}),_jsx(_components.th,{children:`Con`}),_jsx(_components.th,{children:`Recommended`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{style:{textAlign:`center`},children:`Small (≤ 500)`}),_jsx(_components.td,{children:`Precise search`}),_jsx(_components.td,{children:`Context may be cut`}),_jsx(_components.td,{children:`FAQ, short paragraphs`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{style:{textAlign:`center`},children:`Medium (1000)`}),_jsx(_components.td,{children:`Balanced performance`}),_jsx(_components.td,{children:`—`}),_jsx(_components.td,{children:_jsx(_components.strong,{children:`Most cases (default)`})})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{style:{textAlign:`center`},children:`Large (≥ 2000)`}),_jsx(_components.td,{children:`Wider context preserved`}),_jsx(_components.td,{children:`Lower search precision`}),_jsx(_components.td,{children:`Long narrative documents`})]})]})]})}),_jsx(Accordion,{title:`Why is Chunk Overlap needed?`,icon:`circle-question`,id:`why-is-chunk-overlap-needed`,children:_jsx(_components.p,{children:`When sentences are cut at chunk boundaries, related content may be missed during search. Overlap shares some text between adjacent chunks to preserve context. The default 100 is appropriate in most cases.`})})]}),`\n`,_jsx(Note,{children:_jsxs(_components.p,{children:[`Enabling `,_jsx(_components.strong,{children:`Bypass Embedding and Retrieval`}),` skips text splitting and embedding, injecting the entire document directly into the LLM context. Use only for small documents — large documents may exceed token limits.`]})}),`\n`,_jsx(Heading,{level:`3`,id:`semantic-chunking`,children:`Semantic Chunking`}),`\n`,_jsx(Info,{children:_jsxs(_components.p,{children:[_jsx(_components.strong,{children:`New feature`}),` — A method that splits chunks based on `,_jsx(_components.strong,{children:`inter-sentence meaning similarity`}),` instead of fixed size.`]})}),`\n`,_jsxs(_components.p,{children:[`When Text Splitter is set to `,_jsx(_components.strong,{children:`Semantic`}),`, sentences are converted to embedding vectors, and chunk boundaries are auto-determined where similarity between adjacent sentences drops sharply.`]}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{}),_jsx(_components.th,{children:`Fixed Size (Character/Token)`}),_jsx(_components.th,{children:`Semantic`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Split criterion`})}),_jsx(_components.td,{children:`Character / token count`}),_jsx(_components.td,{children:`Meaning similarity change`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Pro`})}),_jsx(_components.td,{children:`Fast and predictable`}),_jsx(_components.td,{children:`Splits at topic boundaries, improves search accuracy`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Con`})}),_jsx(_components.td,{children:`May cut mid-sentence/paragraph`}),_jsx(_components.td,{children:`Requires embedding engine, longer processing`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Recommended`})}),_jsx(_components.td,{children:`General documents`}),_jsx(_components.td,{children:`Long documents with frequent topic shifts`})]})]})]}),`\n`,_jsx(Warning,{children:_jsx(_components.p,{children:`Semantic chunking requires an embedding engine to be configured. Without one, picking Semantic causes errors.`})}),`\n`,_jsx(Heading,{level:`3`,id:`table-preservation`,children:`Table Preservation`}),`\n`,_jsx(Info,{children:_jsxs(_components.p,{children:[_jsx(_components.strong,{children:`New feature`}),` — Preserve tables (HTML/markdown format) in documents intact instead of splitting.`]})}),`\n`,_jsxs(_components.p,{children:[`Standard chunking can split tables across multiple chunks, losing row/column information. With `,_jsx(_components.strong,{children:`Table Preservation`}),` enabled:`]}),`\n`,_jsxs(_components.ol,{children:[`\n`,_jsxs(_components.li,{children:[`Auto-detect tables in the document (HTML `,_jsx(_components.code,{children:`\u003ctable>`}),`, markdown `,_jsx(_components.code,{children:`|...|`}),`)`]}),`\n`,_jsx(_components.li,{children:`Standard chunking on text portions only`}),`\n`,_jsx(_components.li,{children:_jsx(_components.strong,{children:`Tables are attached intact to the closest text chunk`})}),`\n`,_jsx(_components.li,{children:`Tables exceeding chunk size are split row-by-row while preserving the header`}),`\n`]}),`\n`,_jsxs(_components.p,{children:[`Controlled by the `,_jsx(_components.strong,{children:`Preserve Tables`}),` option in profile advanced settings (default: enabled).`]}),`\n`,_jsx(Tip,{children:_jsx(_components.p,{children:`Especially effective for table-heavy financial reports and technical specifications. Search accuracy for specific cell values within tables is greatly improved.`})}),`\n`,_jsx(Heading,{level:`3`,id:`contextual-chunking`,children:`Contextual Chunking`}),`\n`,_jsx(Info,{children:_jsxs(_components.p,{children:[_jsx(_components.strong,{children:`New feature`}),` — Each chunk has a context summary of the entire document generated by the LLM and prepended. Implements Anthropic’s Contextual Retrieval technique.`]})}),`\n`,_jsx(_components.p,{children:`In later chunks of long documents, earlier context is lost, causing search matching failures. With contextual preservation enabled:`}),`\n`,_jsxs(_components.ol,{children:[`\n`,_jsx(_components.li,{children:`After chunking completes, call LLM for each chunk`}),`\n`,_jsx(_components.li,{children:`Generate a summary of “where this chunk is in the entire document and its context”`}),`\n`,_jsx(_components.li,{children:`Prepend the summary to the chunk before vectorization`}),`\n`]}),`\n`,_jsx(CodeBlock,{numberOfLines:`11`,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:`11`,children:[_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:`[Before contextual preservation]`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:` Chunk: \"In conclusion, the initial hypothesis was confirmed.\"`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:` → Can't tell what \"A hypothesis\" refers to → search fails`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:`[After contextual preservation]`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:` Chunk: \"This chunk is the conclusion of the A Hypothesis Verification Report,`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:` containing the final judgment integrating 50 experiment results.`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:` In conclusion, the initial hypothesis was confirmed.\"`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{children:` → Includes context → search succeeds`})}),`\n`,_jsx(_components.span,{className:`line`,children:_jsx(_components.span,{})})]})})}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Setting`}),_jsx(_components.th,{children:`Description`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Contextual Chunking`})}),_jsx(_components.td,{children:`Activate/deactivate toggle`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Model`})}),_jsx(_components.td,{children:`LLM model used for context summarization`})]})]})]}),`\n`,_jsx(Warning,{children:_jsxs(_components.p,{children:[_jsx(_components.strong,{children:`LLM calls happen per chunk count.`}),` A document with 100 chunks → 100 LLM calls. Bulk uploading large documents can greatly increase API costs. Use selectively for small or important documents.`]})}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`embedding-settings`,children:`Embedding Settings`}),`\n`,_jsx(_components.p,{children:`Configure the embedding engine and model that converts documents to vectors.`}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Setting`}),_jsx(_components.th,{style:{textAlign:`center`},children:`Default`}),_jsx(_components.th,{children:`Description`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Embedding Model Engine`})}),_jsx(_components.td,{style:{textAlign:`center`},children:`SentenceTransformers`}),_jsx(_components.td,{children:`Service for vector conversion`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Embedding Model`})}),_jsx(_components.td,{style:{textAlign:`center`},children:_jsx(_components.code,{children:`all-MiniLM-L6-v2`})}),_jsx(_components.td,{children:`Model name to use`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Batch Size`})}),_jsx(_components.td,{\"data-numeric\":`true`,style:{textAlign:`center`},children:`32`}),_jsx(_components.td,{children:`Documents processed per batch (only shown for Ollama, OpenAI, Azure)`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Embedding Dimension`})}),_jsx(_components.td,{style:{textAlign:`center`},children:`0 (auto)`}),_jsx(_components.td,{children:`Vector dimension count. 0 uses model default`})]})]})]}),`\n`,_jsx(Warning,{children:_jsxs(_components.p,{children:[`Changing the embedding model makes existing document vectors incompatible. After changing the model, `,_jsx(_components.strong,{children:`all Knowledge Bases must be reindexed`}),`.`]})}),`\n`,_jsx(_components.p,{children:_jsx(_components.strong,{children:`Supported Engines:`})}),`\n`,_jsxs(Tabs,{children:[_jsxs(Tab,{title:`SentenceTransformers`,id:`sentencetransformers`,children:[_jsx(_components.p,{children:`Open-source embedding engine running locally.`}),_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Setting`}),_jsx(_components.th,{children:`Description`})]})}),_jsx(_components.tbody,{children:_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Model`})}),_jsxs(_components.td,{children:[`HuggingFace model name (e.g., `,_jsx(_components.code,{children:`sentence-transformers/all-MiniLM-L6-v2`}),`)`]})]})})]})]}),_jsx(Tab,{title:`OpenAI`,id:`openai`,children:_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Setting`}),_jsx(_components.th,{children:`Description`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`API URL`})}),_jsx(_components.td,{children:`OpenAI API Endpoint`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`API Key`})}),_jsx(_components.td,{children:`API authentication key`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Model`})}),_jsxs(_components.td,{children:[`e.g., `,_jsx(_components.code,{children:`text-embedding-3-small`})]})]})]})]})}),_jsx(Tab,{title:`Azure OpenAI`,id:`azure-openai`,children:_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Setting`}),_jsx(_components.th,{children:`Description`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`API URL`})}),_jsx(_components.td,{children:`Azure OpenAI Endpoint`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`API Key`})}),_jsx(_components.td,{children:`Azure auth key`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`API Version`})}),_jsx(_components.td,{children:`API version`})]})]})]})}),_jsx(Tab,{title:`Ollama`,id:`ollama`,children:_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Setting`}),_jsx(_components.th,{children:`Description`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`API URL`})}),_jsx(_components.td,{children:`Ollama server address`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Model`})}),_jsxs(_components.td,{children:[`e.g., `,_jsx(_components.code,{children:`nomic-embed-text`})]})]})]})]})}),_jsx(Tab,{title:`Vertex AI`,id:`vertex-ai`,children:_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Setting`}),_jsx(_components.th,{children:`Description`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Project ID`})}),_jsx(_components.td,{children:`Google Cloud project ID`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Location`})}),_jsxs(_components.td,{children:[`Region (default: `,_jsx(_components.code,{children:`us-central1`}),`)`]})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Service Account Key`})}),_jsx(_components.td,{children:`Global key or Custom JSON key`})]})]})]})}),_jsxs(Tab,{title:`Gemini`,id:`gemini`,children:[_jsx(_components.p,{children:`Uses Google Gemini embedding API.`}),_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Setting`}),_jsx(_components.th,{children:`Description`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`API Key`})}),_jsx(_components.td,{children:`Gemini API auth key`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Model`})}),_jsxs(_components.td,{children:[`Embedding model name (e.g., `,_jsx(_components.code,{children:`text-embedding-004`}),`)`]})]})]})]})]})]}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`file-upload-limits`,children:`File Upload Limits`}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Setting`}),_jsx(_components.th,{style:{textAlign:`center`},children:`Default`}),_jsx(_components.th,{children:`Description`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Max file size`})}),_jsx(_components.td,{style:{textAlign:`center`},children:`Unlimited`}),_jsx(_components.td,{children:`Single file upload max size (in MB)`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Max files`})}),_jsx(_components.td,{style:{textAlign:`center`},children:`Unlimited`}),_jsx(_components.td,{children:`Files usable simultaneously per chat`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Allowed file extensions`})}),_jsx(_components.td,{style:{textAlign:`center`},children:`All allowed`}),_jsxs(_components.td,{children:[`Comma-separated (e.g., `,_jsx(_components.code,{children:`pdf, docx, txt`}),`). Empty = all extensions allowed`]})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`PDF conversion extensions`})}),_jsx(_components.td,{style:{textAlign:`center`},children:`Disabled`}),_jsxs(_components.td,{children:[`Extensions to convert to PDF via LibreOffice (e.g., `,_jsx(_components.code,{children:`pptx, docx, xlsx`}),`). LibreOffice installation required`]})]})]})]}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`question-generation`,children:`Question Generation`}),`\n`,_jsx(_components.p,{children:`A feature that pre-generates “questions a user might ask” for each chunk via LLM to improve search accuracy.`}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Setting`}),_jsx(_components.th,{style:{textAlign:`center`},children:`Default`}),_jsx(_components.th,{children:`Description`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Enabled`})}),_jsx(_components.td,{style:{textAlign:`center`},children:`OFF`}),_jsx(_components.td,{children:`Toggle on`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Model`})}),_jsx(_components.td,{style:{textAlign:`center`},children:`—`}),_jsx(_components.td,{children:`LLM for question generation`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Max questions per chunk`})}),_jsx(_components.td,{\"data-numeric\":`true`,style:{textAlign:`center`},children:`10`}),_jsx(_components.td,{children:`Range 1–20`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Question vector weight`})}),_jsx(_components.td,{\"data-numeric\":`true`,style:{textAlign:`center`},children:`0.5`}),_jsx(_components.td,{children:`0.0 (content only) ~ 1.0 (questions only). 0.5 = equal weight`})]})]})]}),`\n`,_jsx(Warning,{children:_jsx(_components.p,{children:`Enabling adds LLM calls during document processing. Increases processing time and cost — use only when search accuracy is insufficient.`})}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`cloud-storage-integration`,children:`Cloud Storage Integration`}),`\n`,_jsx(_components.p,{children:`Import documents into Knowledge Bases from external cloud storage.`}),`\n`,_jsxs(_components.ul,{children:[`\n`,_jsxs(_components.li,{children:[`This setting only provides toggles — `,_jsx(_components.strong,{children:`actual auth info must be set via environment variables`}),`.`]}),`\n`]}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Storage`}),_jsx(_components.th,{style:{textAlign:`center`},children:`Toggle`}),_jsx(_components.th,{children:`Required Environment Variables`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Google Drive`})}),_jsx(_components.td,{style:{textAlign:`center`},children:`ON/OFF`}),_jsxs(_components.td,{children:[_jsx(_components.code,{children:`GOOGLE_DRIVE_CLIENT_ID`}),`, `,_jsx(_components.code,{children:`GOOGLE_DRIVE_API_KEY`})]})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`OneDrive`})}),_jsx(_components.td,{style:{textAlign:`center`},children:`ON/OFF`}),_jsx(_components.td,{children:_jsx(_components.code,{children:`ONEDRIVE_CLIENT_ID`})})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`SharePoint`})}),_jsx(_components.td,{style:{textAlign:`center`},children:`ON/OFF`}),_jsxs(_components.td,{children:[_jsx(_components.code,{children:`ONEDRIVE_CLIENT_ID_BUSINESS`}),`, `,_jsx(_components.code,{children:`SHAREPOINT_TENANT_ID`}),`, `,_jsx(_components.code,{children:`SHAREPOINT_SITE_URL`})]})]})]})]}),`\n`,_jsx(Note,{children:_jsxs(_components.p,{children:[`Environment variables must be set to enable toggles. Enabling a toggle shows the cloud source in the Knowledge Base’s `,_jsx(_components.strong,{children:`Add Content`}),` menu.`]})}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`reindexing-and-reset`,children:`Reindexing and Reset`}),`\n`,_jsx(Danger,{children:_jsxs(_components.p,{children:[`The actions below are `,_jsx(_components.strong,{children:`irreversible`}),`. Always confirm before executing.`]})}),`\n`,_jsxs(Table,{children:[_jsx(_components.thead,{children:_jsxs(_components.tr,{children:[_jsx(_components.th,{children:`Action`}),_jsx(_components.th,{children:`Description`}),_jsx(_components.th,{children:`Impact Scope`})]})}),_jsxs(_components.tbody,{children:[_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Reindex Knowledge Bases`})}),_jsx(_components.td,{children:`Rebuild vector indexes for all KBs`}),_jsx(_components.td,{children:`All Knowledge Bases (takes time)`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Reset Vector Storage`})}),_jsx(_components.td,{children:`Delete all data in the Vector DB`}),_jsx(_components.td,{children:`Deletes all vectors → reindexing required`})]}),_jsxs(_components.tr,{children:[_jsx(_components.td,{children:_jsx(_components.strong,{children:`Reset Upload Directory`})}),_jsx(_components.td,{children:`Delete all files uploaded to the server`}),_jsx(_components.td,{children:`Deletes original files`})]})]})]}),`\n`,_jsx(Heading,{level:`3`,id:`when-reindexing-is-needed`,children:`When Reindexing is Needed`}),`\n`,_jsxs(_components.ul,{children:[`\n`,_jsx(_components.li,{children:`When you’ve changed the embedding model`}),`\n`,_jsx(_components.li,{children:`When you’ve changed the search engine (Vector DB)`}),`\n`,_jsx(_components.li,{children:`When you’ve changed chunk size/overlap`}),`\n`,_jsx(_components.li,{children:`When document processing has issues`}),`\n`]}),`\n`,_jsx(_components.hr,{}),`\n`,_jsx(Heading,{level:`2`,id:`related-pages`,children:`Related Pages`}),`\n`,_jsxs(Columns,{cols:3,children:[_jsx(Card,{title:`Knowledge Base`,icon:`database`,href:`/en/workspace/knowledge`,children:_jsx(_components.p,{children:`Create Knowledge Bases and manage documents — pick profile per KB`})}),_jsx(Card,{title:`Search Engine`,icon:`magnifying-glass`,href:`/en/admin/settings/search-engine`,children:_jsx(_components.p,{children:`Vector DB and search parameter settings`})}),_jsx(Card,{title:`Dynamic Filters`,icon:`filter`,href:`/en/workspace/knowledge-filters`,children:_jsx(_components.p,{children:`Knowledge Base metadata filter settings`})})]})]})}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":"Documents","description":"Configure document processing profiles, content extraction, embedding engines, and search parameters.","href":"/en/admin/settings/documents"}}},"mdxSourceWithNoJs":null},"mdxExtracts":{"tableOfContents":[{"title":"Extraction Engine","slug":"extraction-engine","depth":2,"children":[]},{"title":"Document Processing Profiles","slug":"document-processing-profiles","depth":2,"children":[{"title":"What is a Profile?","slug":"what-is-a-profile","depth":3,"children":[]},{"title":"Profile Management","slug":"profile-management","depth":3,"children":[]}]},{"title":"Content Extraction","slug":"content-extraction","depth":2,"children":[{"title":"LLM Vision Extraction","slug":"llm-vision-extraction","depth":3,"children":[]}]},{"title":"Text Splitting","slug":"text-splitting","depth":2,"children":[{"title":"Semantic Chunking","slug":"semantic-chunking","depth":3,"children":[]},{"title":"Table Preservation","slug":"table-preservation","depth":3,"children":[]},{"title":"Contextual Chunking","slug":"contextual-chunking","depth":3,"children":[]}]},{"title":"Embedding Settings","slug":"embedding-settings","depth":2,"children":[]},{"title":"File Upload Limits","slug":"file-upload-limits","depth":2,"children":[]},{"title":"Question Generation","slug":"question-generation","depth":2,"children":[]},{"title":"Cloud Storage Integration","slug":"cloud-storage-integration","depth":2,"children":[]},{"title":"Reindexing and Reset","slug":"reindexing-and-reset","depth":2,"children":[{"title":"When Reindexing is Needed","slug":"when-reindexing-is-needed","depth":3,"children":[]}]},{"title":"Related Pages","slug":"related-pages","depth":2,"children":[]}],"codeExamples":{}},"jsLinks":["/_mintlify/custom/cloocus/45db6ee3c57396cb/toc.js","/_mintlify/custom/cloocus/9670a2c81dc38d5e/tools/playwright-video-mcp/index.js","/_mintlify/custom/cloocus/4540e1c308555295/lang-toggle.js"],"subdomain":"cloocus","siteName":"Cloosphere Guide","title":"Documents - Cloosphere Guide","navMaps":{"firstHrefInVersion":{},"firstHrefInLanguage":{"ko":{"href":"/ko/admin/settings/documents","matchLevel":1},"en":{"href":"/en/admin/settings/documents","matchLevel":0}}},"lastModified":"2026-08-21T07:39:12.783Z","originalFileLocation":"en/admin/settings/documents.mdx","locale":"en"},"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=\"Configure document processing profiles, content extraction, embedding engines, and search parameters.\">\u003cmeta data-mint-head name=\"application-name\" content=\"Cloosphere Guide\">\u003cmeta data-mint-head name=\"generator\" content=\"Mintlify\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"192x192\" href=\"/mintlify-assets/_mintlify/favicons/cloocus/WlRjtofBr-j1DZPc/_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/cloocus/WlRjtofBr-j1DZPc/_generated/favicon/apple-touch-icon.png\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"16x16\" href=\"/mintlify-assets/_mintlify/favicons/cloocus/WlRjtofBr-j1DZPc/_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/cloocus/WlRjtofBr-j1DZPc/_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/cloocus/WlRjtofBr-j1DZPc/_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/cloocus/WlRjtofBr-j1DZPc/_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/cloocus/WlRjtofBr-j1DZPc/_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/cloocus/WlRjtofBr-j1DZPc/_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/cloocus/WlRjtofBr-j1DZPc/_generated/favicon-dark/favicon.ico\" media=\"(prefers-color-scheme: dark)\">\u003cmeta data-mint-head property=\"og:title\" content=\"Documents - Cloosphere Guide\">\u003cmeta data-mint-head property=\"og:description\" content=\"Configure document processing profiles, content extraction, embedding engines, and search parameters.\">\u003cmeta data-mint-head property=\"og:site_name\" content=\"Cloosphere Guide\">\u003cmeta data-mint-head property=\"og:url\" content=\"https://docs.cloosphere.ai/en/admin/settings/documents\">\u003cmeta data-mint-head property=\"og:type\" content=\"website\">\u003cmeta data-mint-head property=\"og:image\" content=\"https://cloocus.mintlify.site/_mintlify/api/og/v2?division=Knowledge+%26+Documents\u0026amp;title=Documents\u0026amp;description=Configure+document+processing+profiles%2C+content+extraction%2C+embedding+engines%2C+and+search+parameters.\u0026amp;theme=97296152c031fdcf5b54d0af\u0026amp;signature=624c43bdda44737b147e3b965abdab4aac7988cc00af944123376f728c5a1e3b\">\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=\"Documents - Cloosphere Guide\">\u003cmeta data-mint-head name=\"twitter:description\" content=\"Configure document processing profiles, content extraction, embedding engines, and search parameters.\">\u003cmeta data-mint-head name=\"twitter:image\" content=\"https://cloocus.mintlify.site/_mintlify/api/og/v2?division=Knowledge+%26+Documents\u0026amp;title=Documents\u0026amp;description=Configure+document+processing+profiles%2C+content+extraction%2C+embedding+engines%2C+and+search+parameters.\u0026amp;theme=97296152c031fdcf5b54d0af\u0026amp;signature=624c43bdda44737b147e3b965abdab4aac7988cc00af944123376f728c5a1e3b\">\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/cloocus/WlRjtofBr-j1DZPc/_generated/favicon/browserconfig.xml\">\u003cmeta data-mint-head name=\"apple-mobile-web-app-title\" content=\"Cloosphere Guide\">\u003cmeta data-mint-head name=\"msapplication-TileColor\" content=\"#5f22d0\">\u003clink data-mint-head rel=\"canonical\" href=\"https://docs.cloosphere.ai/en/admin/settings/documents\">\u003clink data-mint-head rel=\"alternate\" type=\"application/xml\" href=\"/sitemap.xml\">\u003clink data-mint-head rel=\"alternate\" type=\"text/markdown\" href=\"/en/admin/settings/documents.md\">\u003clink data-mint-head rel=\"alternate\" hreflang=\"en\" href=\"https://docs.cloosphere.ai/en/admin/settings/documents\">\u003clink data-mint-head rel=\"alternate\" hreflang=\"ko\" href=\"https://docs.cloosphere.ai/ko/admin/settings/documents\">\u003clink data-mint-head rel=\"alternate\" hreflang=\"x-default\" href=\"https://docs.cloosphere.ai/ko/admin/settings/documents\">\u003cscript data-mint-head type=\"application/ld+json\">{\"@context\":\"https://schema.org\",\"@graph\":[{\"@type\":\"Organization\",\"@id\":\"https://docs.cloosphere.ai/#organization\",\"name\":\"Cloosphere Guide\",\"url\":\"https://docs.cloosphere.ai\",\"logo\":{\"@type\":\"ImageObject\",\"url\":\"https://mintcdn.com/cloocus/uXAlvLAzCJlwDYsB/images/logo.png?fit=max\u0026auto=format\u0026n=uXAlvLAzCJlwDYsB\u0026q=85\u0026s=93fb26230c58b4ba9411c3e97e20b439\"}},{\"@type\":\"WebSite\",\"@id\":\"https://docs.cloosphere.ai#website\",\"name\":\"Cloosphere Guide\",\"url\":\"https://docs.cloosphere.ai\",\"publisher\":{\"@id\":\"https://docs.cloosphere.ai/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https://docs.cloosphere.ai/en/admin/settings/documents#webpage\",\"url\":\"https://docs.cloosphere.ai/en/admin/settings/documents\",\"name\":\"Documents\",\"description\":\"Configure document processing profiles, content extraction, embedding engines, and search parameters.\",\"dateModified\":\"2026-08-21T07:39:12.783Z\",\"isPartOf\":{\"@id\":\"https://docs.cloosphere.ai#website\"},\"breadcrumb\":{\"@id\":\"https://docs.cloosphere.ai/en/admin/settings/documents#breadcrumb\"}},{\"@type\":\"BreadcrumbList\",\"@id\":\"https://docs.cloosphere.ai/en/admin/settings/documents#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Settings\",\"item\":\"https://docs.cloosphere.ai/en/admin/settings/general\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Documents\",\"item\":\"https://docs.cloosphere.ai/en/admin/settings/documents\"}]},{\"@type\":[\"Article\",\"TechArticle\"],\"@id\":\"https://docs.cloosphere.ai/en/admin/settings/documents#article\",\"headline\":\"Documents\",\"name\":\"Documents\",\"description\":\"Configure document processing profiles, content extraction, embedding engines, and search parameters.\",\"url\":\"https://docs.cloosphere.ai/en/admin/settings/documents\",\"mainEntityOfPage\":{\"@id\":\"https://docs.cloosphere.ai/en/admin/settings/documents#webpage\"},\"dateModified\":\"2026-08-21T07:39:12.783Z\",\"publisher\":{\"@id\":\"https://docs.cloosphere.ai/#organization\"},\"isPartOf\":{\"@id\":\"https://docs.cloosphere.ai#website\"}}]}\u003c/script>\u003cscript data-mint-head type=\"application/ld+json\">{\"@context\":\"https://schema.org\",\"@type\":\"WebSite\",\"name\":\"Cloosphere Guide\",\"publisher\":{\"@type\":\"Organization\",\"name\":\"Cloosphere Guide\"},\"provider\":{\"@type\":\"Organization\",\"name\":\"Mintlify\",\"url\":\"https://mintlify.com\"}}\u003c/script>","agentPointerHref":"/en/admin/settings/documents.md","siteName":"Cloosphere Guide","nav":{"global":null,"languages":[{"language":"ko","default":true,"pages":[{"title":"ko","href":"/ko/admin/settings/documents"}]},{"language":"en","tabs":[{"tab":"Getting Started","icon":"rocket","pages":[{"title":"Getting Started","href":"/en"}]},{"tab":"Chat","icon":"comments","pages":[{"title":"Chat","href":"/en/chat/overview"}]},{"tab":"Workspace","icon":"briefcase","pages":[{"title":"Workspace","href":"/en/workspace/overview"}]},{"tab":"Collaboration","icon":"users","pages":[{"title":"Collaboration","href":"/en/collaboration/overview"}]},{"tab":"Automation","icon":"clock","pages":[{"title":"Automation","href":"/en/automation/overview"}]},{"tab":"Admin","icon":"gear","groups":[{"group":"Admin","pages":[{"title":"Admin Panel","href":"/en/admin/overview"}]},{"group":"Users","pages":[{"title":"User Management","href":"/en/admin/users"},{"title":"Groups \u0026 Permissions","href":"/en/admin/groups"},{"title":"Organization Management","href":"/en/admin/organizations"},{"title":"Inquiries","href":"/en/admin/inquiries"}]},{"group":"Settings","pages":[{"group":"General \u0026 Connections","expanded":false,"pages":[{"title":"General","href":"/en/admin/settings/general"},{"title":"Connections","href":"/en/admin/settings/connections"},{"title":"Models","href":"/en/admin/settings/models"},{"title":"Marketplace","description":"Connect the service containers for PPT/Word/Excel/HTML document generators and Google Workspace/Microsoft 365 integrations.","href":"/en/admin/settings/marketplace"}]},{"group":"Knowledge \u0026 Documents","expanded":false,"pages":[{"title":"Documents","href":"/en/admin/settings/documents"},{"title":"Document Templates","description":"Register master PowerPoint · Word · Excel templates that agents use when generating documents.","href":"/en/admin/settings/document-templates"},{"title":"Search Engine","href":"/en/admin/settings/search-engine"},{"title":"Web Search","href":"/en/admin/settings/web-search"}]},{"group":"AI Features \u0026 Safety","expanded":false,"pages":[{"title":"Guardrails","href":"/en/admin/settings/guardrails"},{"title":"Code Execution","href":"/en/admin/settings/code-execution"},{"title":"Code Gateway","href":"/en/admin/code-gateway"},{"title":"Audio","href":"/en/admin/settings/audio"},{"title":"Images","href":"/en/admin/settings/images"}]},{"group":"UI \u0026 Channels","expanded":false,"pages":[{"title":"Interface","href":"/en/admin/settings/interface"},{"title":"Branding","href":"/en/admin/settings/branding"},{"title":"Channels","href":"/en/admin/notifications"},{"title":"Microsoft Teams Bot","href":"/en/admin/teams-bot"},{"title":"Embed Widgets","href":"/en/admin/settings/embed-widgets"}]},{"group":"Data \u0026 Operations","expanded":false,"pages":[{"title":"Data Retention","href":"/en/admin/settings/data-retention"},{"title":"Memory","href":"/en/admin/settings/memory"},{"title":"Monitoring","href":"/en/admin/settings/monitoring"},{"title":"Database","href":"/en/admin/settings/database"},{"title":"Encryption","href":"/en/admin/settings/encryption"},{"title":"License","href":"/en/admin/settings/license"}]}]}]},{"tab":"Monitoring","icon":"chart-line","pages":[{"title":"Monitoring","href":"/en/monitoring/overview"}]},{"tab":"Resources","icon":"screwdriver-wrench","pages":[{"title":"Resources","href":"/en/admin/deployment-checklist"}]}]}]},"basePath":"","title":"Documents - Cloosphere Guide","deploymentId":"dpl_DyXGnKeCjZyA3sUdfMJ7FjkdBefg","buildId":"dpl_DyXGnKeCjZyA3sUdfMJ7FjkdBefg","dynamics":["07c022d89b","a74be98ba7","a9bb46feb3"],"chunks":["Frame-D0FT0qQs.js","DynamicMarkdown-CQM9Bhbv.js","zoom-image-lightbox-BQ4tbj_9.js","lightbox-handle-r17bVUz3.js","Markdown-CX2QY19L.js","AssistantPageLink-CY3bFglh.js","dist-Dl6_zDnT.js","lib-DMeTj1gb.js","lib-B8HbPgcC.js","katex-Dbtb6Wk7.js","html-void-elements-mCSPHezU.js","marked.esm-CZWythHk.js"]}