Nutrient Java SDK
Need pricing or implementation help? Talk to Sales.
WORD TO PDF
import io.nutrient.sdk.Document;
try (Document document = Document.open("report.docx")) { // Convert Word → PDF in one call document.exportAsPdf("output.pdf"); // Also: Excel, PowerPoint, HTML, Markdown}Three engines — OCR, ICR, and VLM-enhanced ICR — extract text, tables, equations, and key-value pairs from scanned and digital documents. Run on-premises or connect to Claude, OpenAI, or local VLMs.
Convert between PDF, Word, Excel, PowerPoint, HTML, and Markdown. Produce accessible PDF/UA output. No external dependencies required.
Add annotations, form fields, digital signatures, and stamps. Redact sensitive content with permanent removal. Manage pages, merge documents, and edit metadata.
Generate documents dynamically from Word templates with data-driven content. Convert template output to accessible PDF/UA for compliance workflows.
Extract structured data from any document using OCR, ICR, or VLM-enhanced processing.
Convert between PDF, Office formats, HTML, and Markdown with high-fidelity output.
Add annotations, text, stamps, and shapes to PDFs. Manage pages, merge files, and edit metadata.
Create and fill interactive form fields. Apply visible and invisible digital signatures.
Permanently remove sensitive content from documents for compliance and privacy.
Generate documents dynamically from Word templates with data-driven content.
Every document processing capability in one SDK. AI extraction runs on-premises by default, with optional cloud VLM enhancement. Editing, conversion, and generation run entirely locally.
VISION API
Vision API is the AI core of the Nutrient Java SDK. Run intelligent content recognition entirely on-premises, or enhance accuracy with Claude, OpenAI, or local vision language models. Extract tables, equations, key-value pairs, and document structure — or generate natural language image descriptions for accessibility and cataloging.
OCR and ICR run entirely on your servers with zero external calls. Meets HIPAA, GDPR, and air-gapped requirements out of the box.
Connect to Claude, OpenAI, or local models (Ollama, LM Studio, vLLM) for the highest accuracy on complex financial, legal, and medical documents.
Every extraction returns classified elements with bounding boxes, confidence scores, and hierarchical reading order. Ready for programmatic processing.
Generate WCAG-compliant alt text and contextual descriptions from document images. Choose your VLM provider based on privacy and accuracy needs.
The SDK covers the full document lifecycle. Beyond AI-powered extraction (OCR, ICR, VLM-enhanced), it includes document conversion between PDF, Word, Excel, PowerPoint, HTML, and Markdown. It provides PDF editing with annotations, form fields, digital signatures, and content redaction, and it supports template-based document generation from Word templates with accessible PDF/UA output. All capabilities work on-premises with a single dependency.
The SDK converts Word (DOCX) to PDF and PDF to Word, Excel (XLSX) to PDF and PDF to Excel, PowerPoint (PPTX) to PDF and PDF to PowerPoint, Markdown to PDF, and PDF to HTML. It also supports Word-to-PDF/UA conversion for accessible document output. All conversions run locally without external services.
OCR is the fastest engine, extracting text with word-level bounding boxes. ICR (intelligent content recognition) is an AI-powered engine that runs on-premises — it detects tables, equations, key-value regions, and document structure without external API calls. VLM-enhanced ICR adds a vision language model (Claude, OpenAI, or local) on top for the highest accuracy on complex documents. All three return structured JSON output.
Yes. All SDK capabilities — extraction, conversion, editing, and generation — run on your infrastructure. The OCR and ICR engines require zero external API calls. VLM-enhanced mode can also stay on-premises when connected to a local model server (Ollama, LM Studio, or vLLM). Document conversion, PDF editing, and template generation are entirely local with no cloud dependencies.
Google Document AI and AWS Textract are cloud-only extraction services — your documents must be uploaded to their servers. Nutrient Java SDK runs on your infrastructure by default and offers much more than extraction alone. It combines AI-powered data extraction with document conversion, PDF editing, digital signatures, redaction, and template generation in a single dependency. You get data sovereignty, predictable costs, and a complete document processing platform — not just an extraction API.
The SDK supports eight annotation types (text, free text, shapes, stamps, sticky notes, links, text markup, and redaction), form field creation and filling, visible and invisible digital signatures with advanced workflows, page management (reordering, merging, adding custom pages), and metadata editing. Redaction annotations permanently remove content from the document, supporting GDPR and HIPAA compliance.
Yes. The SDK processes Word templates with dynamic content injection, enabling you to generate reports, contracts, and other documents programmatically from data. Template output can be converted directly to accessible PDF/UA format for compliance workflows. This is useful for automated document generation pipelines where you need consistent formatting with variable data.
VLM-enhanced ICR supports Anthropic Claude, OpenAI, and any OpenAI-compatible custom endpoint. The custom endpoint option works with Ollama, LM Studio, vLLM, and other local inference servers. You can switch providers with a single configuration change. Image description also supports all three provider types, giving you full control over accuracy, cost, and data privacy.
Yes. On-premises processing means documents never leave your servers. Content redaction permanently removes sensitive data. Digital signatures provide document integrity and authentication. PDF/UA output meets accessibility standards. The combination of on-premises extraction, redaction, and signatures makes the SDK suitable for healthcare (HIPAA), finance (SOC 2), legal, and government (air-gapped) environments.
Add Nutrient Java SDK as a dependency to your project and follow the getting started guide. All capabilities — extraction, conversion, editing, and generation — are available immediately. The documentation includes step-by-step guides for each feature area with working examples you can adapt for your use case.