Document AI
Parse, extract, and process structured data from documents.
Tool | Category | Segment | Platform / Tool | Plan | Monthly Price USD | Pricing Model | Free Tier / Trial | Included Usage / Limits | Document / OCR Capabilities | File Types / Modalities | Extraction / RAG / LLM Features | API / SDK / Integrations | Storage / Retention | Deployment / Hosting | Team / Governance | Best Fit | Main Limits / Caveats |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
No tagline | Document AI | AI-native document parsing | LlamaParse / LlamaCloud | Free | $0 | Credit-based monthly cloud plan | ✓ | Official homepage states 10,000 free credits/month, roughly 1,000 pages, with agentic OCR, schema extraction and document agents | Agentic OCR and document parsing optimized for AI/RAG | 50+ unstructured file types, including complex PDFs, images and handwritten notes per product page | Parse, Extract, Split, Classify and Index features for document agents | LlamaParse API/client and LlamaIndex/LlamaCloud integrations | Cloud handling under LlamaIndex account; privacy/docs per LlamaCloud | LlamaCloud hosted service | Free account; enterprise for SSO/VPC/hybrid | Developers building RAG over complex PDFs without implementing parsers | Credit/page cost varies by parsing mode; official detailed pricing can be hard to inspect without account |
No tagline | Document AI | Cloud document OCR / IDP | Google Cloud Document AI | Enterprise Document OCR Processor | $1.50/1K pages first 5M; $0.60/1K pages after | Per-page processor pricing | New Google Cloud customers get $300 cloud credit | Enterprise OCR Processor: $1.50 per 1,000 pages for 1-5M pages/month; OCR add-ons $6 per 1,000 pages | OCR, handwriting, layout-aware document text digitization | PDF, images and supported processor input formats; page counted by file/page rules | OCR output can feed Document AI Layout Parser, Vertex AI Search, RAG and downstream extraction | Google Cloud APIs, client libraries, Workbench/console and processor endpoints | Google Cloud data handling, processors and storage depend on project configuration | Google Cloud managed service | IAM, audit logging, VPC/security controls through Google Cloud | Teams already on GCP needing scalable OCR for documents | No permanent Document AI free tier was visible on pricing page; other Google Cloud services may add costs |
No tagline | Document AI | Cloud document OCR / IDP | Google Cloud Document AI | Layout Parser / Form Parser / Custom Extractor | $10-$30/1K pages | Per-page processor pricing | New Google Cloud customers get $300 cloud credit | Layout Parser $10/1K pages; Custom Extractor and Form Parser $30/1K pages for first 1M pages/month | Layout parsing, form parsing, custom entity extraction and specialized processors | Document pages, forms, PDFs and supported document formats | Layout Parser includes initial chunking; extractors produce structured entities for RAG/workflows | Google Cloud API and processor endpoints | Cloud project storage/retention depends on processor and app setup | Google Cloud managed processors | IAM, service accounts, audit logs and enterprise controls | Structured extraction from forms, invoices, contracts or domain documents | Per-page extractor cost is much higher than simple OCR; custom processor deployment can add operational cost |
No tagline | Document AI | Cloud document OCR / IDP | Amazon Textract | Free tier | $0 for first 3 months within quota | AWS Free Tier page quotas | Yes, for new AWS customers | 3 months: Detect Document Text 1,000 pages/month; Analyze Document 100 pages/month for Forms/Tables/Layout and query combos; Expense and ID 100 pages/month; Lending 2,000 pages/month | OCR, handwriting, tables, forms, signatures, queries, expenses, IDs and lending docs | Documents and images supported by Textract APIs | Structured extraction from tables/forms/queries; expense and ID-specific extraction | AWS SDKs, CLI, APIs, Lambda/S3/event workflows | Data storage depends on app/S3 usage; async APIs use job outputs | AWS managed service by region | IAM, CloudTrail, VPC endpoints where available and AWS compliance controls | AWS teams testing OCR and IDP without upfront spend | Free tier lasts only 3 months and excludes Custom Queries |
No tagline | Document AI | Cloud document OCR / IDP | Amazon Textract | Detect Document Text | $1.50/1K pages first 1M in US West example | Per-page pay-as-you-go | Free tier exists for first 3 months | Official pricing example: $0.0015/page for first 1M pages in US West (Oregon); $0.0006/page after 1M in example | OCR text and handwriting extraction | Documents and images supported by Textract | Raw text extraction for downstream search/RAG | AWS SDKs, CLI and APIs | Application controlled; output can be stored in S3 | AWS managed regional service | AWS IAM and account governance | High-volume OCR where AWS integration matters | Pricing is regional and feature-specific; examples are not a substitute for calculator |
No tagline | Document AI | Cloud document OCR / IDP | Amazon Textract | Analyze Document / Expense / ID / Lending | Usage-based by feature | Per-page feature pricing | Free tier exists for first 3 months | Examples: Tables $0.015/page, Forms $0.05/page, Queries $0.015/page, Expense $0.01/page, ID $0.025/page, Lending $0.07/page in US West examples | Tables, forms, queries, expense, ID and lending extraction | Forms, tax docs, invoices, IDs, mortgage/lending docs and other document images/PDFs | Feature-specific structured extraction with OCR included in Analyze output | AWS APIs and SDKs; integrates with S3/Lambda/Step Functions | Output retention and storage controlled by application/AWS resources | AWS managed regional service | AWS IAM, monitoring and enterprise account controls | Production IDP workflows on AWS | Feature combinations can get expensive quickly; Custom Queries has no free tier |
No tagline | Document AI | Cloud document OCR / IDP | Azure AI Document Intelligence | Free F0 | $0 | Free monthly page quota | ✓ | F0 supports all Document Intelligence features for testing, with 0-500 pages free per month on pricing page | Read, Layout, prebuilt, custom classification/extraction and add-ons depending feature availability | Documents/images accepted by Azure Document Intelligence APIs; page-based billing | Prebuilt models for documents, receipts, invoices, ID, tax forms, contracts and query/add-on features | REST API, SDKs, Document Intelligence Studio and Azure integrations | Azure service data handling and region controls | Azure managed service; container option shown on pricing page | Azure RBAC, networking, compliance and enterprise controls | Prototyping document extraction on Azure | Free tier is for testing and has rate/volume limits; paid S0 pricing is region-specific |
No tagline | Document AI | Cloud document OCR / IDP | Azure AI Document Intelligence | Standard S0 | Region-specific PAYG | Per-page pay-as-you-go | F0 free tier exists | Paid page pricing varies by feature and region; pricing page lists Read, Layout/prebuilt, custom extraction/classification, query fields, batch and training dimensions | Read OCR, layout, prebuilt models, custom models and batch processing | PDFs/images and supported document formats | Structured extraction, query fields, classification and custom extraction | REST API, SDKs, Studio and Azure service integrations | Azure data handling by region/resource | Azure managed service or container where available | Azure enterprise governance and networking | Production extraction workloads in Azure environments | Official page may render regional prices dynamically; use Azure pricing calculator for final SKU numbers |
No tagline | Document AI | Document OCR API | Mistral OCR | mistral-ocr-latest / OCR 3 | $2/1K pages | Per-page OCR pricing | No free OCR tier captured on pricing page | Pricing page lists OCR 3 at $2 per 1,000 pages and annotations at $3 per 1,000 pages | OCR and document understanding with markdown, tables, images, layout and confidence scores | PDF, image URL and document URL inputs; docs mention PDFs, images, PPTX/DOCX and more | Extracts text while preserving hierarchy; can output tables, headers/footers, images and document annotations | Mistral OCR API, SDKs and batch inference | Mistral platform file/document handling | Mistral hosted API; batch mode for scale | Enterprise terms and data controls depend on Mistral account/contract | Low-cost OCR for LLM-ready markdown from complex documents | OCR pricing separate from LLM token pricing; no free quota found in official pricing |
No tagline | Document AI | Document OCR API | Mistral OCR | OCR with annotations | $3/1K pages annotations | Per-page annotation add-on | No free OCR tier captured | Annotations priced separately from OCR on pricing page; docs expose document and bbox annotation formats | OCR with bounding boxes, page/word confidence and structured annotation output | PDFs, images and supported document inputs | JSON/schema-style annotations for structured document outputs | Mistral OCR endpoint and SDKs | Platform document/file handling | Mistral hosted API; batch inference recommended for scale | Enterprise controls by contract | Teams needing OCR plus structured annotation for VLM/document datasets | Annotation cost stacks with OCR; validate output schema needs before scaling |
No tagline | Document AI | AI-native document parsing | LlamaParse / LlamaCloud | Starter / Pro | $50/mo Starter; $500/mo Pro | Credit subscription plus PAYG | Free plan exists | Common pricing profile: Starter 40K credits/month, Pro 400K credits/month, with pay-as-you-go credit top-ups; verify current checkout before buying | Higher-volume LlamaParse, LlamaExtract and LlamaCloud document workflows | Complex PDFs, images, tables, charts, forms and multimodal documents | Schema extraction, document agents, indexing and retrieval workflows | API/client, LlamaIndex framework and cloud workflows | LlamaCloud account/project retention and privacy controls | LlamaCloud managed platform | Paid plans add users/support; enterprise adds SSO/VPC/hybrid | Production document-agent teams using LlamaIndex | Credit-to-page mapping depends on parsing mode; verify account dashboard for exact rates |
No tagline | Document AI | Unstructured document processing | Unstructured | Free | $0 | One-time/free page allowance | ✓ | Pricing page lists 15,000 free pages with no expiration | Partitions and cleans unstructured documents for GenAI/RAG | PDFs, Office docs, images and many unstructured formats depending pipeline | Document partitioning, chunking, cleaning and RAG-ready outputs | API, SDKs, platform workflows and open-source library | Cloud/API data handling; enterprise dedicated/VPC options | Hosted API/platform or self-managed open-source library | Dedicated instance/VPC and multi-user access on enterprise | Teams preparing messy enterprise docs for RAG | Free library and paid API differ in quality/features; enterprise pricing is sales-led |
No tagline | Document AI | Unstructured document processing | Unstructured | Dedicated / Enterprise | Custom | Sales-led dedicated/VPC pricing | Free pages exist | Dedicated instance or VPC with multi-user access, full data isolation, support and tailored pricing | Production document parsing and preprocessing for GenAI at scale | Enterprise file formats and unstructured document corpora | Chunking, cleaning, extraction and data prep for retrieval and agent pipelines | API/platform workflows and enterprise deployments | Dedicated data isolation and custom deployment controls | Dedicated cloud instance or VPC | Multi-user access, data isolation and dedicated support | Enterprise RAG pipelines with private document corpora | No public unit price for dedicated plans; must scope with sales |
No tagline | Document AI | OCR/workflow automation API | Nanonets | Starter | $0 entry with $200 credits | Run/block-based credit pricing | ✓ | Pricing page: Start free with $200 in credits; no platform fees; up to 3 users; data extraction AI, API access, email integration and cloud storage connectors | Data extraction AI for invoices, receipts and document workflows | Invoices, receipts, emails, files and connected storage workflows | OCR/extraction blocks and workflow automation | API access, email integration, cloud storage connectors | Cloud storage connectors and platform handling | Nanonets cloud automation platform | Up to 3 users on Starter; Growth/Enterprise for larger teams | Testing document automation without platform fee | Cost depends on number of workflow runs and block prices; pricing calculator/account needed for exact unit costs |
No tagline | Document AI | OCR/workflow automation API | Nanonets | Growth / Enterprise | Custom / volume pricing | Quote-based volume pricing | Starter credits exist | Growth adds classification AI, barcode/signature detection, generative AI blocks, Python blocks, ERP/database integrations and up to 40% volume discount; Enterprise custom | Document extraction plus end-to-end automation workflows | Invoices, receipts, forms and business documents | Classification, extraction, generative blocks and custom automations | API, email, ERP, database and custom integrations | Platform storage/connectors; enterprise compliance options | Nanonets hosted platform | Growth up to larger teams; Enterprise for compliance/deployment requirements | High-volume AP/ops teams automating document workflows | Quote-based pricing reduces public cost transparency |
No tagline | Document AI | OCR API / document extraction | Mindee | Starter | EUR 44/mo annual billing | Monthly credit subscription | Free trial available | 500 credits/month billed annually; additional credits EUR 0.05; unlimited models; community support | OCR APIs for invoices, receipts, bank statements, IDs and custom models | Page-based physical documents regardless of type/file format | Pretrained and custom document extraction; confidence scores and polygons on higher tiers | Mindee OCR APIs and integrations | Data processing localization shown in plan comparison by tier | Mindee hosted API | Members/support increase by tier; Enterprise custom SLA/support | Predictable low-volume OCR API usage | EUR annual billing; advanced RAG/features start higher |
No tagline | Document AI | OCR API / document extraction | Mindee | Pro / Business | EUR 179/mo Pro; EUR 584/mo Business annual billing | Monthly credit subscription plus overage | Free trial available | Pro: 2,500 credits/month and RAG for 20 documents; Business: 10,000 credits/month and unlimited RAG; overages EUR 0.04/EUR 0.035 per credit | OCR/document extraction for standard and custom document types | Physical pages across document types and file formats | RAG, polygons, confidence scores, boosted accuracy and priority support by tier | API integrations and workflow access options | Data processing localization and enterprise options by tier | Mindee hosted API | Priority support and Enterprise custom SLAs | Teams that need OCR plus RAG/document-question workflows | Annual billing and per-page credits; enterprise needed for custom volume/SLA |
No tagline | Document AI | Receipt/invoice OCR API | Veryfi | Free | $0 | Monthly document quota | ✓ | Pricing page: process up to 100 docs/month free; all document types, SDKs for development, limited storage, email support | Multi-modal OCR/data extraction for invoices, receipts and business documents | Invoices, receipts, checks and other supported document types | Line-item extraction, OCR 3.0, document capture SDK and data extraction APIs | Veryfi OCR API, SDKs and docs | Limited storage on Free; Vault/custom retention on higher tiers | Veryfi hosted platform and SDKs | Email support on Free; Growth adds SAML/SLA/custom retention | Developers testing invoice/receipt extraction API | Free limit is 100 docs/month; storage and support limited |
No tagline | Document AI | Receipt/invoice OCR API | Veryfi | Starter / Growth | $500/mo minimum Starter; Growth custom | Transaction-based API pricing | Free plan exists | Starter minimum $500/mo buying roughly <5K docs/month; FAQ lists receipt $0.08 and invoice $0.16 in Starter; Growth volume discounts and custom terms | OCR/data extraction APIs plus SDKs, fraud detection and document capture add-ons | Invoices, receipts, checks, purchase orders and other supported docs | Line items, extraction, product matching/workflows on higher tiers | API Hub, SDKs, OpenClaw Skill and add-ons | Limited storage on Starter; Growth has Vault, unlimited storage and custom retention | Veryfi hosted API/platform | Growth adds Slack support, SAML SSO, SLA options, model training | Finance/AP teams needing fast receipt/invoice OCR | Starter has a high monthly minimum; add-ons may increase price |
No tagline | Document AI | PDF services / extraction API | Adobe Acrobat Services API | Free Tier | $0 | Document transactions per month | ✓ | 500 free Document Transactions per month; access to 15+ PDF Services including PDF Extract, Auto-Tag, Electronic Seal and Document Generation; no credit card | PDF extraction, generation, conversion, accessibility tagging and PDF workflows | PDF and document service inputs/outputs supported by Acrobat Services APIs | PDF Extract can extract text/tables/structure for downstream apps/RAG | Adobe PDF Services API and SDKs | Adobe service data handling and transaction limits | Adobe cloud API | Adobe developer credentials; paid plans/support for volume | Developers needing free monthly PDF extraction/conversion quota | Not a full OCR/IDP suite; transaction accounting varies by operation/output |
No tagline | Document AI | PDF services / extraction API | Adobe Acrobat Services API | Paid Plans | Custom / sales | Volume and multi-product discounts | Free tier exists | Paid plans provide scalable high-volume access to 15+ PDF Services and technical support on certain plans | High-volume PDF extraction/generation/conversion/auto-tag workflows | PDF and supported document transformations | Document generation and extract workflows for apps | Adobe APIs and SDKs | Adobe cloud service handling | Adobe managed API | Support available on certain paid plans; enterprise procurement | Companies embedding PDF APIs into production software | Public page does not show self-serve per-transaction paid price |
No tagline | Document AI | Rule-based document parser | Docparser | 14-day free trial | $0 trial | Parsing-credit subscription after trial | ✓ | 14-day free trial, no credit card required; 1 parsing credit equals 1 document with up to 5 pages | PDF/Word/image parsing with parser templates and rules | PDF, Word and image files | Extract fields/tables and export structured data | Google Sheets export plus many integrations; downloads to Excel, CSV, JSON and XML | Document retention add-on available | Docparser hosted service | Teams/managed users on Professional+; MFA/version control add-ons | Trying template/rule-based parsing before subscribing | Trial only; complex layouts may need paid parsing assistant/setup |
No tagline | Document AI | Rule-based document parser | Docparser | Starter / Business | $39/mo Starter monthly; $159/mo Business monthly | Monthly parsing credits | 14-day trial | Starter monthly: 100 parsing credits/month and up to 15 parsers; Business monthly: 1,000 parsing credits/month, 500 parsers, priority support and multi-layout parsers | Template/rule-based document extraction | PDF, Word and image files | Smart checkboxes/tables, multi-layout parsers and parser version control by tier | Google Sheets, CSV, JSON, XML and hundreds of integrations | Extended document retention is paid add-on or enterprise feature | Docparser cloud | Teams/managed users and MFA/version control by tier | Operations teams with repeatable document templates | Credit is document up to 5 pages; add-ons can materially change cost |
No tagline | Document AI | Math/scientific OCR API | Mathpix | Convert API | Usage-based; no API free trial | API conversion pricing | No Convert API free trial; Snip app has free plan | Official API pricing page says no free trial for Convert API; Snip app can be used to try capabilities | OCR and conversion for math, STEM, PDFs and structured formats | Images, PDFs and math/scientific documents | LaTeX/math OCR, PDF conversion and structured outputs | Mathpix APIs and SDK workflows | Platform/API account handling | Mathpix hosted API | Account/team controls depend on product plan | Scientific PDFs, equations and STEM document conversion | No permanent free API tier captured; source page should be checked for exact endpoint rates |
No tagline | Document AI | Document Q&A / PDF AI | Humata | Free | $0 | Monthly free pages | ✓ | 60 free pages monthly; 1 user; basic features | Chat with PDFs/documents and answer questions from sources | PDF pages and documents uploaded to Humata | Document Q&A with cited context; OCR starts on Team tier per plan table | Web app and plan-based workflow; API not emphasized in pricing page | Humata account/cloud storage | Humata hosted app | Single user; higher tiers add team/security | Students/researchers chatting with small PDFs | No OCR on Free according to pricing table; only 60 pages/month |
No tagline | Document AI | Document Q&A / PDF AI | Humata | Expert / Team | $9.99/mo Expert; $49/user/mo Team | Subscription plus additional page usage | Free plan exists | Expert: 500 free pages/month, 3 users, additional pages $0.02/page; Team: 5,000 pages/month, 10 users, additional pages $0.01/page and OCR/security features | PDF/document Q&A and OCR on Team tier | PDFs/documents uploaded to Humata | GPT-5 support, OCR, response personalization and permissions by tier | Web app workflow; integrations not primary | Cloud account storage/pages | Humata hosted app | Team adds department/folder permissions; Enterprise adds SOC 2/SLA | Small teams doing document research and PDF Q&A | Page overages add cost; OCR only appears at Team tier |
No tagline | Document AI | Notebook/document research assistant | NotebookLM | Free | $0 | Consumer/product usage limits | ✓ | Google help says users can sign up free; limits include 3 Audio Overviews/day on free tier in upgrade table; sources/notebooks limits are subject to change | Grounded research assistant over uploaded sources | Docs, PDFs, websites, Google docs/slides and other supported NotebookLM sources | Cited answers, summaries, study guides, Audio Overviews and source-grounded Q&A | Web app; Google Workspace/AI plan integrations for upgraded access | Google account data handling; Workspace/enterprise terms vary | Google hosted product | Upgraded plans through Google AI, Cloud or Workspace; enterprise/admin controls by plan | Personal research, study and document synthesis without API integration | Not an OCR/API product; limits change and official page says usage limits are subject to change |
No tagline | Document AI | Notebook/document research assistant | NotebookLM | Plus / Pro / Ultra via Google plans | Varies by Google AI/Workspace/Cloud plan | Bundled subscription access | Free plan exists | Upgrade page lists higher limits and features through Google AI Plans, Google Cloud or qualifying Workspace plans; Audio Overviews examples include 6/day, 20/day and 200/day tiers | Higher-capacity document research assistant | Uploaded/linked sources supported by NotebookLM | More output generation, higher limits and collaboration/controls depending plan | Google product integrations rather than standalone API | Google account/Workspace/Cloud policies | Google hosted product | Workspace/Cloud can add admin controls | Organizations that want NotebookLM workflows with higher limits | Pricing is tied to broader Google AI/Workspace products, not a standalone page-based OCR API |
No tagline | Document AI | Open-source document parser | Docling | Open source | $0 software | Open-source software; infra/model costs separate | ✓ | No software usage meter; install with pip, run CLI/library locally; Docling Serve and Docling MCP available | Converts messy documents into structured data with tables, formulas, reading order, OCR and chunking | PDF, DOCX, PPTX, XLSX, HTML, images, audio transcripts and other formats listed on site | Exports JSON, Markdown, HTML, text and chunks for AI/RAG/agent systems | Python library, CLI, Docling Serve and MCP | Local/app-owned unless using external OCR/models/services | Local, self-hosted or your own infrastructure | Governance depends on deployment; enterprise support not inherent to OSS | Private/local document conversion for RAG pipelines | You own scaling, OCR engine choice and quality tuning |
No tagline | Document AI | Open-source document converter | Microsoft MarkItDown | Open source | $0 software | Open-source library; optional external service costs | ✓ | No software usage meter; converts files/office docs to Markdown for LLM ingestion | Lightweight document-to-Markdown conversion preserving important structure | Local files, remote URIs and byte streams; Office docs, PDFs and other formats via plugins/dependencies | Markdown output for RAG, prompt context and AI ingestion | Python package/CLI; optional integrations such as Azure Document Intelligence for some conversions | Local unless remote URI/service integrations are used | Local/self-hosted | Governance depends on your environment | Simple file-to-Markdown conversion in AI pipelines | Not a full OCR/IDP platform; quality varies by file type and optional services |
No tagline | Document AI | Open-source PDF/document extraction | MinerU | Open source | $0 software | Open-source document parsing engine; infra/model costs separate | ✓ | No software usage meter; converts complex documents like PDFs and Office docs into LLM-ready Markdown/JSON | High-accuracy document parsing and layout/content extraction | PDFs, images and Office docs per ecosystem docs | Markdown/JSON for LLM pretraining, RAG and agentic workflows | CLI, SDK ecosystem and open-source repository | Local/app-owned unless cloud/API services are used | Local/self-hosted or via ecosystem API if chosen | Governance depends on deployment | Research/document-heavy RAG pipelines needing open-source extraction | Requires local setup/resources; production support and SLAs are self-managed |
No tagline | Document AI | Open-source OCR/document extraction | Datalab Marker | Open source / platform | $0 software for OSS; hosted/platform options may vary | Open-source models plus platform offerings | Yes for OSS | Datalab page describes open-source models for extracting text, tables, images and layouts with OCR in 90+ languages | Advanced OCR and document conversion to structured outputs | PDFs, Office documents and images | Text, tables, images, layouts and GitHub Markdown table conversion | Open-source tooling plus Datalab platform/API options | Local/app-owned for OSS; platform data handling if hosted | Local/self-hosted for OSS or Datalab platform | Governance depends on chosen deployment | Developers needing OCR/table extraction from PDFs into Markdown/JSON | Hosted pricing not captured here; OSS requires GPU/ops for best performance |
No tagline | Document AI | Open-source OCR | PaddleOCR | Open source | $0 software | Open-source OCR toolkit; infra/model costs separate | ✓ | No software usage meter; open-source OCR models and pipelines | OCR, layout/document understanding and multilingual text recognition depending model/pipeline | Images, documents and OCR datasets/workflows | Text extraction and document understanding for downstream pipelines | Python ecosystem, models and deployment options | Local/app-owned | Local/self-hosted/cloud by user | Governance depends on deployment | Teams needing a mature open-source OCR baseline | Requires engineering and model selection; not a managed extraction API |
No tagline | Document AI | Open-source RAG / document Q&A | AnythingLLM | Open source | $0 software | Open-source app; model/vector/hosting costs separate | ✓ | No software usage meter for self-hosted app; all-in-one AI app with RAG and agent capabilities | Document ingestion and knowledge-base chat | Uploaded documents and data sources supported by AnythingLLM | RAG, agents and chat over documents/data | Web app, integrations and model/provider connectors | Self-hosted or cloud account handling depending deployment | Self-hosted/local/cloud by user | Governance depends on deployment and edition | Teams wanting a ready document-chat app over private docs | Not an OCR parser by itself; quality depends on document ingestion and chosen models |