Daniel van Strien

@danielvanstrien.bsky.social

4669 Followers 2421 Following

Machine Learning Librarian at @hf.co

input image and ocr output side by side

demo screenshot showing search results and play buttons next to transcripts

slopegraph showing flips in ratings

Horizontal bar chart titled "Agents calling the Hugging Face Hub", showing each coding agent's share of agent-attributed huggingface_hub requests for June 2026. claude-code leads at 23.9%, followed by codex 19.8%, cursor-cli 10.2%, antigravity 2.7%, openclaw 2.3%, hermes-agent 1.4%, then pi, opencode, github-copilot and cursor below 1%. A gray bar at the bottom shows 37.7% of traffic as "unknown" — unregistered tools. Source: hf.co/datasets/huggingface/agent-usage.

Screenshot of duckdb query and an output table ranking content types in common crawl as page counts and pcts

Diff between boto3 client using buckets and S3 showing two new lines for enpoint_url and config

Screen recording of an app showing before after ocr and layour predictions from the model

Side by side: on the left, a scanned 1901 newspaper page (The Commoner); on the right, the clean text Surya OCR 2 read from it — the masthead and opening editorial, in reading order.

A 1901 newspaper page overlaid with Surya OCR 2's detected layout: colored boxes around each block (text, section-header, picture) and numbered dots joined by a path showing the model's reading order across the columns.

A scanned archival page showing nine typewritten index cards arranged in a 3×3 grid. Each card has been detected by the model and outlined with a green bounding box, with confidence scores between 0.93 and 0.99.

Screenshot of the overview of the lineage app

Two archival index cards side by side. Left, labelled "blank": a bare card with only a punch hole near the bottom centre. Right, labelled "content (not-blank)": a typed catalogue card with shelf number, book title, date, and a library form stamp.

Language breakdown

A worked example of NuExtract3 turning a document image into structured data. Top: a terminal command — hf jobs uv run --image vllm/vllm-openai nuextract3.py index-cards cards-json --template schema.json. Middle (input): a scanned typewritten library index card reading "ABAD (Joseph), Captain, Spanish Army, letter of (1783), 5538, f.11." Bottom (output): the JSON extracted from the card — image_type "index_card", heading "Abad J.", heading_type "person", epithet "Captain, Spanish Army", and one entry with ms_no "5538", folios ["f.11"], description "letter of (1783)".

Screenshot of an Embedding Atlas visualization showing 2 million Open Library book titles as a scatter plot. Points are colored by subject category — History (orange), Fiction (green), Religion (purple), Law & Politics (red), and others. Clusters are labeled with topic keywords like "Church-Christian-Bible-God", "Shakespeare", "surgery-anatomy-physiology". A tooltip shows a selected book titled "Rosabelle" categorized as Society, published 1995. The right panel shows a category distribution bar chart.

Night-vision camera trap image of a white-tailed deer in a forest. The deer is highlighted with a red segmentation mask overlay showing pixel-level object boundaries, with a green bounding box and "deer 96%" confidence label. Generated automatically by SAM3 from a text prompt with zero training data.

Leaderboard race visualisation showing OCR models improving benchmark scores over time

Interactive Embedding Atlas visualization of 250,000 training examples from NVIDIA's Nemotron post-training v3 collection, colored by category. Distinct clusters are visible for Math (blue, 56k), Code (orange, 45k), Agentic (green, 41k), Instruction Following (red, 25k), Finance (purple, 18k), Multilingual (brown, 18k), Science (pink, 18k), Safety (grey, 16k), and Identity (yellow, 8k). A selected point shows an example from the Agentic function-calling dataset: "I need to know the weight of the creature that can evolve into a Fire Dragon."

Screenshot of Mirador IIIF viewer showing a panoramic map of Amesville, Athens County Ohio from 1875, served from a Hugging Face Storage Bucket. The map shows detailed bird's-eye view illustrations of buildings, streets, and surrounding farmland.

Code screenshot show loading from the hub filtering and pushing to a bucket

Screenshot of plot showing ELO vs paramter count for different OCR models

Screenshot of a search UI showing a text box with search results showing index cards next to the ocr for the card

Image with historic newspaper on the left and output from OCR models sampled on the right. The output on the right shows okay starting text and then a lot of repetition.

Screenshot of old vs new ocr.

old ocr text is garbled. New ocr much cleaner.

table of contents showing ocr models supported in the repo

Screen recording show a user searching for "medical" and "medical segmentation" using keyword and semantic search. Results show arxiv abstracts.

image of a index card with a green bounding box prediction around the card contents

Screenshot of the application. shows a user choosing a demo iiif manifest and clicking classify. the application shows sparklines and predictions for which pages contain illustrations.

Here's alt text for the meme:

Alt text: "Flex Tape meme format. Top panel: Phil Swift (labeled 'Library Systems Vendor') aggressively spraying water representing 'Outdated systems, metadata issues, disjointed search and complex user needs.' Bottom panel: A hand slapping Flex Tape underwater, with the tape labeled 'AI-powered chat interface.'"

This captures the joke that vendors are positioning AI chat as a quick fix for deep-seated library infrastructure problems—a bit like slapping tape on a leak rather than fixing the plumbing.

Screenshot of a simple app showing bounding boxes for photographs detected in historic newspaper images.

Meme showing two versions of the doge dog.
Top: muscular buff doge labeled 'Public Domain circa 2000' saying 'Open access to knowledge and culture is a universal good'.
Bottom: small weak doge labeled 'Public Domain post LLMs?' saying 'Someone might use public domain content for training an LLM' with worried emoji.

screenshot of latex output

$hf jobs uv run --flavor a100-large --timeout 2h \ -s HF_TOKEN \ https://huggingface.co/datasets/uv-scripts/ocr/raw/main/deepseek-ocr-vllm.py \ NationalLibraryOfScotland/Britain-and-UK-Handbooks-Dataset \ davanstrien/handbooks-deep-ocr \ --resolution-mode base \ --batch-size 2048 \ --prompt-mode free$

Logs showing ocr progress

Video of animated squares floating in the screen and circling around. One becomes the whole screen and the text "Small models for GLAM" appears.

screenshot of the hf jobs command to run the model.

Screenshot of an index card with annotated bounding box predictions from the ocr model

Picture of a digitised index card.

Screenshot of the dataset viewer showing a column of marc data + the first few pages of an open access monograph

Screenshot of the iconclass-vlm model demo showing predictions for a 17th century portrait painting of a standing woman in black dress with white ruff collar. The interface displays the model's raw JSON prediction with ICONCLASS codes, then compares predictions against ground truth labels in two columns. Model correctly identifies "31A231 standing figure" and "61B(+55) historical persons (portraits and scenes from the life) (+ full length portrait)" among others, achieving 3 out of 6 matches. Some predictions marked as "Not a valid iconclass label" showing areas where the model needs improvement.

Screenshot of a hf jobs uv run command with some flags and a URL pointing to a script.

Screenshot of an app showing an image from a page + model reasoning showing how the model is parsing the text and layout.

Screenshot of a plyabill with some OCR results on the right

Screenshot of the app showing a page from a book + different views of existing and new ocr.

Document page with text

A recording showing a long list of authors who contributed to the dataset

Screenshot showing a document page image on the left with corresponding OCR output on the right of the page.

Heritage organisations should actively develop open-source AI models to provide alternatives to big tech’s control over cultural data.

Screenshot of the header of the article with text:

AI Scraping Bots Are Breaking Open Libraries, Archives, and Museums

Sure grandma let's get you to bed meme with text

I manually wrote XSLT to convert ALTO XML OCR into AI-ready text

Sure grandma, let's get you to bed.

screenshot of token

Screen recording of the Claude Desktop app using the mcp server to find non maths reasoning datasets via a few different tools.

Screenshot of this text:

Meanwhile, we distilled the chain-of-thought from DeepSeek-R1-0528 to post-train Qwen3 8B Base, obtaining DeepSeek-R1-0528-Qwen3-8B. This model achieves state-of-the-art (SOTA) performance among open-source models on the AIME 2024, surpassing Qwen3 8B by +10.0% and matching the performance of Qwen3-235B-thinking. We believe that the chain-of-thought from DeepSeek-R1-0528 will hold significant importance for both academic research on reasoning models and industrial development focused on small-scale models.

+ a table of metrics

Screenshot of the dataset viewer on the Hugging Face Hub. Shows a set of metadata for the newspaper navigator dataset. It also has previews of a few rows showing images alongside metadata columns.

Image of a historic newspaper with bounding box predictions for "photographs" "headline" "illustration" etc.

Flex Tape meme showing a person labeled 'LIBRARY SYSTEMS VENDOR' trying to fix major problems (labeled 'OUTDATED SYSTEMS, METADATA ISSUES, DISJOINTED SEARCH AND COMPLEX USER NEEDS') with an 'AI-POWERED CHAT INTERFACE' represented by the Flex Tape product.