AI LEGAL ASSISTANTS SHOULDN’T SILENTLY REWRITE YOUR DOCUMENTS
@nutrientdocs built Document Authoring AI to keep counsel in control.
→ every AI edit lands as a tracked change, not a silent overwrite
→ each change comes with a comment explaining why it was made
→ tested it on a real NDA in the browser, full DOCX fidelity preserved
→ export the reviewed doc straight to DOCX when you’re done
counsel reviews suggestions instead of trusting a black box. that’s the difference
demo → nutrient.io/sdk/document-aut…
Nutrient retweeted
OCR reads every character. Document AI reads meaning. Most production pipelines need both, routed through a validation layer that catches hallucinations before they hit your database. Here's the framework for picking the right tier per document.
Nutrient retweeted
In all seriousness, we @nutrientdocs trained grounding models that score how faithful each extracted value is to the source, and put them on HF with an open leaderboard + a benchmark anyone can rerun.
On number-grounding the open English model is ~0.92; general-purpose NLI baselines sit ~0.45–0.66.
huggingface.co/nutrientdocs/…
Nutrient retweeted
We're making steady progress climbing the ParseBench leaderboard. Latest changes made Nutrient jump from 45th to 12th, ahead of most dedicated parsing products and LLMs, just behind Fable. At a fraction of the cost of running a frontier model over every page.
Nutrient retweeted
Replying to @nutrientdocs
@nutrientdocs here with what you’ll see at #TXD26: We embed document workflows directly into Salesforce.
Think about generating, editing, redacting, annotating, and signing natively with no external tools and no data leaving your org.
See it live in Austin at booth 3!
Nutrient retweeted
#TXD26, generate documents straight from Salesforce data — merge fields, repeating sections, conditions, related records.
No queries required. We’ll walk you through @nutrientdocs at booth 3!
Nutrient retweeted
Quick brag: @nutrientdocs has a habit of winning Demo Jams. 😎
We'll defend our title from #LondonsCalling at #TXD26: three minutes, one Salesforce-native document demo.
Come judge for yourself.
Nutrient retweeted
Bring your toughest Salesforce document problem to @nutrientdocs at #TXD26 booth 3.
We’ll solve it.
Signatures, form fields, annotations, page restructuring, redaction — all without leaving Salesforce.
Nutrient retweeted
For regulated work, an Agent that edits the document isn't useful until every change it made is something a human can see, diff, and reverse. The edit is the easy part. The audit trail is the product.
Nutrient retweeted
We've built an on-device AI Assistant for PDFs at @nutrientdocs last year on top of the Foundation Models which actually works and doesn't give up for PDFs with more than 10 pages.
With the pipelines we have built, these updates can be game changing!
I'm so so excited for these!
Nutrient retweeted
Redacting sensitive documents still needs a human in the loop.
Freedom of Information requests demand careful redaction — slow, tedious, human work.
Nutrient PDF Editor pairs Claude Cowork's speed with human review to cut the time without risking mistakes.
See video
Nutrient retweeted
Have you noticed Anthropic's shipped pdf-viewer plugin never actually shows the PDF?
It's fundamentally broken and has been for a while.
So we at Nutrient built one that works.
nutrient.io/claude-desktop
Nutrient retweeted
New 1 step agentic setup just landed. For any platform. getapdfviewer.com
Nutrient retweeted
Is your SDK ready for AI agents? It's free and simple to find out now. Great work by one of my colleagues at @nutrientdocs - where our core product is SDK - and I am so happy he open sourced it :) Let's hope many will follow his example.
Nutrient retweeted
Shipping Day! We’re all proud and excited @nutrientdocs to officially release our PDF to Markdown converter. On our 200 document hand annotated markdown based benchmark it beats out every other major parser for speed and reading order while matching accuracy.
We are shipping it as a closed source, free to use for up to 1,000 documents per month, no license key required, and your documents stay completely local.
We’ve had early beta testers give us phenomenal feedback but please reach out and share publicly your experience with it.
Our future roadmap:
- adding our OCR Vision capabilities to it.
- supporting Windows x64
- speed increases (yeah, make it even faster)
- support for DOCX to md
- more based on feedback.
If you find it useful, please star the repo and tell your friends!
Nutrient retweeted
We've been cooking @nutrientdocs on a PDF, DOCX, and other common document file types converter to Markdown packaged as a simple CLI library. On all our benchmarks we've done so far it's the fastest available by a long shot, most token efficient, and on par with accuracy of the top libraries.
- 2x faster than liteparse that was just released.
- 5x faster than pymupdf
- 4x faster than markit who @badlogicgames just praised for its speed.
- 3x faster than Microsoft's markitdown.
We'll be releasing the converter as freemium closed sourced here soon. Stay tuned.
Let me know if your interested in early access.
Nutrient retweeted
ITDigest’s Weekly News Roundup Featuring @mobileum_ , @Grameenphone, @nutrientdocs, @Domotalk, @veritoneinc, @Oracle, @IBM, @Xero , @AnthropicAI, @SEQSTER, @CrowdStrike and @intel
Read More: itdigest.com/artificial-inte…
#artificialintelligence #Businesstechnology #cloudsecurity
Nutrient retweeted
We're spending a significant amount of time polishing the UX of @nudocsai so it's fast and feels great. Here's a sneak peek of the 'thermometer' widget that helps you navigate any document or skill file.
Nutrient retweeted
I signed the contract today with software my company built years ago. (now it's @nutrientdocs, still best for anything PDF)























