Rewrite org README for the deterministic-document-platform positioning

Replace the AI-Assistant-era card with the current positioning (Nutrient, formerly PSPDFKit, the deterministic document platform for agents), an asset table of every model/dataset/leaderboard/demo in the org with licenses, where each fits in an agentic document workflow, and production entry points (Data Extraction, MCP Server, skills, on-prem commercial models). No benchmark numbers in the README; leaderboards are the source of truth. All links verified live.

README.md

@@ -1,55 +1,66 @@
 ---
 title: README
-emoji: 🦀
-colorFrom: pink
-colorTo: indigo
+emoji: 📊
+colorFrom: gray
+colorTo: green
 sdk: static
 pinned: false
 ---
 
-# Nutrient.io 🚀
+# Nutrient
 
-Welcome to the official Nutrient.io HuggingFace organization! 🌟 We're on a mission to revolutionize the **document space** with cutting-edge AI solutions. Whether you're working with complex PDF workflows, document annotations, or data extraction, our tools and AI assistant are here to streamline and empower your work.
+**Nutrient (formerly PSPDFKit) is the deterministic document platform for agents.** It combines reliable document processing infrastructure for agents, routing and governance, and interfaces for the human in the loop, so that agentic document workflows produce output the person signing off on the business outcome can rely on.
 
----
+This organization is where we publish the small, document-specialized models behind that platform, together with the open benchmarks and self-service leaderboards we score them on. Same task, same data, same scorer, for our models and everyone else's.
 
-## 🌐 About Us
-At [Nutrient.io](https://www.nutrient.io), we specialize in creating tools that make managing and interacting with documents effortless. From **automating repetitive tasks** to **intelligent document understanding**, our solutions are designed to save you time and improve accuracy.
+**At a glance**
 
-Our expertise includes:
-- Advanced **PDF handling** and transformation.
-- Document annotation, extraction, and processing.
-- Easy integration into your existing workflows.
+- Company: Nutrient, Raleigh, North Carolina. Formerly PSPDFKit. Website: [nutrient.io](https://www.nutrient.io)
+- Product entry point for extraction: [Nutrient Data Extraction](https://www.nutrient.io/api/data-extraction-api/) (parse PDFs, scans, images, and Office files; extract schema-defined fields with source citations and per-field confidence signals; free tier, no credit card)
+- What we publish here: task-specific document models (open-weight and commercial), CC-BY-SA / CC-BY / MIT benchmark datasets, reproducible leaderboards, and no-signup demos
+- Release pattern: an open-weight model (Apache-2.0, ONNX where possible) alongside a higher-accuracy commercial model, both scored on the same public benchmark. Commercial weights deploy on-prem, so documents never leave your environment
+- Model-agnostic: everything here works next to any LLM, frontier or local, or none. Our platform overview for agents lives at [nutrient.io/llms.txt](https://www.nutrient.io/llms.txt)
 
----
+## Why small document models
 
-## 🤖 AI Assistant
-The [Nutrient.io AI Assistant](https://www.nutrient.io/sdk/ai-assistant) is designed to help developers build intelligent, context-aware document workflows. Its core capabilities include:  
-- **Context-Aware Prompts**: Generate AI responses based on selected document content, empowering users to interact with documents more effectively.
-- **Dynamic Integrations**: Embed the assistant into document viewers for tailored interactions like summarization, explanation, or answering questions directly from document data.
-- **Offline and On-Premise Models**: Run AI models securely within your environment, ensuring compliance, privacy, and uninterrupted access without relying on external services.
+Most agents can pull data out of a document. Far fewer can prove where it came from, split a scanned batch back into its documents, tell an invoice from a remittance, or find the fields on a form at a quality level that stands up in an audit. Frontier LLMs are strong generalists and weak on exactly these document-shaped tasks: numbers inside tables, long page streams, dropout-ink forms, handwriting next to print.
 
-This functionality empowers you to enhance document-driven applications with powerful, privacy-preserving AI tools.
+We publish narrow models for those tasks and the benchmarks that show where each one wins and where it loses. The point is not to replace your model. It is to give an agentic workflow the deterministic pieces it needs to route, verify, and escalate to a human, and to let you check our numbers yourself.
 
----
+## What's here
 
-## 🚀 Why Nutrient.io?
-- **Developer-Friendly SDKs**: Built to integrate effortlessly into your applications.
-- **Offline & Private Solutions**: Our AI models work entirely on-premise, giving you full control over your data.
-- **Focus on Simplicity**: We believe in making complex workflows accessible to everyone.
+| Task | Open weights | Commercial | Benchmark | Leaderboard | Demo |
+| --- | --- | --- | --- | --- | --- |
+| **Grounding**: does the source document actually support this claim (a number, date, or fact)? | [grounding-en](https://huggingface.co/nutrientdocs/grounding-en) (Apache-2.0) | [grounding-multilingual](https://huggingface.co/nutrientdocs/grounding-multilingual) (15+ languages) | [grounding-benchmark](https://huggingface.co/datasets/nutrientdocs/grounding-benchmark) (CC-BY-SA-4.0) | [grounding-leaderboard](https://huggingface.co/spaces/nutrientdocs/grounding-leaderboard) | [grounding-demo](https://huggingface.co/spaces/nutrientdocs/grounding-demo) |
+| **Document splitting** (page-stream segmentation): where does one document end and the next begin? | [doc-split-v1](https://huggingface.co/nutrientdocs/doc-split-v1) (Apache-2.0, ONNX) | [doc-split-v2](https://huggingface.co/nutrientdocs/doc-split-v2) | [doc-split-benchmark](https://huggingface.co/datasets/nutrientdocs/doc-split-benchmark) (CC-BY-SA-4.0) | [doc-split-leaderboard](https://huggingface.co/spaces/nutrientdocs/doc-split-leaderboard) | [doc-split-demo](https://huggingface.co/spaces/nutrientdocs/doc-split-demo) |
+| **Document classification** (open-vocabulary, zero-shot): classify a page against labels you choose at runtime | [document-classification-v1](https://huggingface.co/nutrientdocs/document-classification-v1) (Apache-2.0, ONNX) | [document-classification-v2](https://huggingface.co/nutrientdocs/document-classification-v2) | [document-classification-benchmark](https://huggingface.co/datasets/nutrientdocs/document-classification-benchmark) | [document-classification-leaderboard](https://huggingface.co/spaces/nutrientdocs/document-classification-leaderboard) | [document-classification-demo](https://huggingface.co/spaces/nutrientdocs/document-classification-demo) |
+| **Document and figure image classification** (open-vocabulary): documents, charts, signatures, screenshots | | [doc-img-classification](https://huggingface.co/nutrientdocs/doc-img-classification) | [doc-openvocab-benchmark](https://huggingface.co/datasets/nutrientdocs/doc-openvocab-benchmark) (CC-BY-SA-4.0) | [doc-openvocab-leaderboard](https://huggingface.co/spaces/nutrientdocs/doc-openvocab-leaderboard) | [doc-openvocab-demo](https://huggingface.co/spaces/nutrientdocs/doc-openvocab-demo) |
+| **Form field detection**: bounding box, type, label, and radio-group identity for every field on a page image | | [form-field-vlm](https://huggingface.co/nutrientdocs/form-field-vlm) | [form-field-vlm-benchmark](https://huggingface.co/datasets/nutrientdocs/form-field-vlm-benchmark) (CC-BY-4.0) | [form-field-vlm-leaderboard](https://huggingface.co/spaces/nutrientdocs/form-field-vlm-leaderboard) | [form-field-vlm-demo](https://huggingface.co/spaces/nutrientdocs/form-field-vlm-demo) |
+| **PII detection and redaction**: context-aware redaction on full documents | | | [DocPII-redaction-benchmark](https://huggingface.co/datasets/nutrientdocs/DocPII-redaction-benchmark) (MIT) | [LLM-PII-Detection-Leaderboard](https://huggingface.co/spaces/nutrientdocs/LLM-PII-Detection-Leaderboard) | |
 
----
+Every leaderboard is scored by the same open scorer we use ourselves. Current numbers, including the cases where a cloud model beats us, live on the leaderboards, not in this README. To add a model, run the scorer and open a pull request on the leaderboard space.
 
-## 🔗 Get Started
-Check out our [documentation](https://www.nutrient.io/sdk/ai-assistant) to learn more and get started with the Nutrient.io AI Assistant.
+## Where these fit in an agentic document workflow
 
----
+- **Extract, then verify.** [Nutrient Data Extraction](https://www.nutrient.io/api/data-extraction-api/) returns each field with a page reference, bounding box, source blocks, a confidence signal, and a match label. The grounding models score whether an extracted value is actually supported by the source, so `fuzzy_match` and `not_found` can be routed to human review while grounded values continue downstream. Confidence is a relative signal, not a calibrated probability, and we say so.
+- **Route and govern.** Splitting and classification turn a mailbox, scanned batch, or upload folder into typed documents with known boundaries, the precondition for [routing, governance, and audit trails](https://www.nutrient.io/document-ai/governance/).
+- **Keep a human in the loop.** Form field detection and the rendered document viewers in our SDKs are the review surface: the human checks the real document, not a transcript of it. See [AI Assistant](https://www.nutrient.io/sdk/ai-assistant/) and [Enterprise Document Agents](https://www.nutrient.io/ai/agents/).
+
+## Use it in production
+
+- **Data Extraction**: REST endpoint, [getting-started guide](https://www.nutrient.io/guides/dws-data-extraction/getting-started/), [citations and confidence](https://www.nutrient.io/guides/dws-data-extraction/extract/citations-and-confidence/), [published benchmarks and methodology](https://www.nutrient.io/api/data-extraction-api/benchmarks/). New accounts get 5,000 credits per month at no cost, no credit card, and signup is agent-compatible.
+- **Interactive extraction demos**, no signup or key: [nutrient-extraction-samples](https://github.com/PSPDFKit/nutrient-extraction-samples). Hover a field, see the exact source region.
+- **Build your own agent**: [MCP Server](https://www.nutrient.io/api/mcp-server-pdf-automation-llm/), SDKs, and Document Engine at [nutrient.io/ai/infrastructure](https://www.nutrient.io/ai/infrastructure/). Agent skills, including [PDF-to-Markdown](https://www.nutrient.io/ai/skills/pdf-to-markdown/) and the local document CLI: [nutrient-skills](https://github.com/PSPDFKit-labs/nutrient-skills).
+- **Commercial models on-prem** or a custom build with our implementation teams: [contact us](https://www.nutrient.io/contact-sales/).
+- **All open models and benchmarks in one place**: [nutrient.io/ai/models](https://www.nutrient.io/ai/models/).
+
+## Citing and reusing
 
-## 💬 Join Us
-We're always looking to collaborate and innovate. Reach out to us with your ideas or questions!
+Datasets carry their own licenses (CC-BY-SA-4.0, CC-BY-4.0, or MIT, stated on each card). Open-weight models are Apache-2.0. If you use a benchmark or leaderboard result, link the leaderboard space so readers can see the scoring code and the full field at the time you cite it.
 
-- 📞 [Contact Sales](https://www.nutrient.io/contact-sales)
-- 🌐 Website: [Nutrient.io](https://www.nutrient.io)
-- 🐦 Twitter: [@nutrientdocs](https://twitter.com/nutrientdocs)
+## Contact
 
-Let’s redefine what’s possible in the document space together! 🚀
+- Website: [nutrient.io](https://www.nutrient.io)
+- GitHub: [PSPDFKit](https://github.com/PSPDFKit) · [PSPDFKit-labs](https://github.com/PSPDFKit-labs)
+- X: [@nutrientdocs](https://x.com/nutrientdocs)
+- Sales and implementation teams: [nutrient.io/contact-sales](https://www.nutrient.io/contact-sales/)