Daniel van Strien @danielvanstrien.bsky.social · 30/09/2026Newspapers from @europeana.bsky.social 070
Daniel van Strien @danielvanstrien.bsky.social · 30/09/2026Uploaded a dataset of 98,877 historical newspaper pages (1700s–1940s) to the Hub, each with its original OCR text, word boxes and confidence scores. huggingface.co/datasets/big... 24716
Daniel van Strien @danielvanstrien.bsky.social · 23/09/2026The Europeana Newspapers dataset on @hf.co now has an `alto` config: the raw ALTO XML for all 5.9M pages. The coordinates for every word, line and block, per-word OCR confidence and font info that the flattened text dropped are back. huggingface.co/datasets/biglam/europeana_newspapershuggingface.cobiglam/europeana_newspapers · Datasets at Hugging FaceWe’re on a journey to advance and democratize artificial intelligence through open source and open science. 1164
Daniel van Strien @danielvanstrien.bsky.social · 16/09/2026Made some improvements to the OCR scripts onboarding in my uv-scripts collection. First run is now a small example: seven scanned NASA pages in, Markdown out, one HF jobs command. It also works on your own images or a folder of PDFs. You can pick from 28 OCR models. huggingface.co/datasets/uv-... 1141
Reposted by Daniel van StrienDaniel van Strien @danielvanstrien.bsky.social · 11/09/2026Need a historical illustration? Search 1.49 million images from British Library books and Britannica (1500s–1920s). Find similar images, download full-resolution images or transparent PNG cutouts, and explore the sources. huggingface.co/spaces/davan... 47620
Daniel van Strien @danielvanstrien.bsky.social · 11/09/2026Need a historical illustration? Search 1.49 million images from British Library books and Britannica (1500s–1920s). Find similar images, download full-resolution images or transparent PNG cutouts, and explore the sources. huggingface.co/spaces/davan... 47620
Daniel van Strien @danielvanstrien.bsky.social · 10/09/2026This includes Kraken’s BLLA segmentation, with raw text output and no table/LaTeX reconstruction. Full results + scripts to reproduce the run: github.com/davanstrien/...github.comocr-bench/experiments/kraken-olmocr-bench at main · davanstrien/ocr-benchPer-collection OCR leaderboards using VLM-as-judge - davanstrien/ocr-bench 040
Daniel van Strien @danielvanstrien.bsky.social · 10/09/2026Used Astra + Jobs to see how well this model performs on the full olmOCR-bench Unsurprisingly, it doesn’t do brilliantly overall: 36.8% But for a ~16M-parameter recogniser, I think 74.4% on long/tiny text and 57.9% on multi-column pages are pretty interesting. 190
Reposted by Daniel van StrienAdina Yakup @adinayakup.bsky.social · 10/09/2026DeepSeek v4.1 Flash is just another level 🤯 huggingface.co/deepseek-ai/... - Asymmetric Causal-Encoder-Decoder: 550B MoE, input 8B / output 16B - Native vision merged into one endpoint - KV cache crushed: ~1/4 the HBM vs last one, 437× smaller than their first modelhuggingface.codeepseek-ai/DeepSeek-V4.1-Flash · Hugging FaceWe’re on a journey to advance and democratize artificial intelligence through open source and open science. 2754
Daniel van Strien @danielvanstrien.bsky.social · 07/09/2026OCR for Japanese manga, Swedish handwriting or Arabic print? There’s a growing range of OCR models on @hf.co: VLMs, dedicated text recognisers and complete OCR pipelines. I’ve gathered 41 models into four collections, with short notes to help you choose: huggingface.co/collections/...huggingface.coOCR on the Hub - a davanstrien CollectionCurated OCR models for documents, languages, handwriting and text in images. Browse four collections with short practical notes. 0288
Daniel van Strien @danielvanstrien.bsky.social · 03/09/2026Fun fact: I had to adjust the plotting logic to get it to show properly because the model does so well! Kraken OCR library: kraken.re/main/index.h... Model: huggingface.co/small-models...kraken.rekraken — kraken documentation 030
Daniel van Strien @danielvanstrien.bsky.social · 03/09/2026A 15.9M-parameter specialist OCR model outperforms many VLMs hundreds of times larger. On a 2,165-page historical eval dataset, Kraken PP-OCRv6 ranks #4 on reading CER and #1 when long-s, ligatures and case are preserved. huggingface.co/spaces/fineb... 2346
Reposted by Daniel van StrienTom Aarsen @tomaarsen.com · 26/08/2026📈 New blog post: Training and Finetuning Multi-Vector Embedding Models with Sentence Transformers. As a practical example, I finetuned a ColBERT-style model for medical retrieval. 14.5 hours on one RTX 3090, and it beats every general-purpose retriever I could find. Thread 🧵 2214
Daniel van Strien @danielvanstrien.bsky.social · 26/08/2026Every illustration in the Encyclopaedia Britannica dataset now has an instance mask: 411,385 cut-outs from 115,293 pages, 1768–1929, each linked to its full-resolution scan. Public domain, no image generation involved. Work in progress. 03710
Daniel van Strien @danielvanstrien.bsky.social · 25/08/2026Uploaded a dataset of 115,293 illustrated pages from the Encyclopaedia Britannica, 1st edition (1768) to 14th (1929) to the Hub huggingface.co/datasets/big... 0335
Daniel van Strien @danielvanstrien.bsky.social · 20/08/2026Dataset (crop_masks config): huggingface.co/datasets/big... Search + cutouts: huggingface.co/spaces/davan... Model: huggingface.co/small-models...huggingface.cobiglam/british-library-book-images · Datasets at Hugging FaceWe’re on a journey to advance and democratize artificial intelligence through open source and open science. 0131
Daniel van Strien @danielvanstrien.bsky.social · 20/08/2026Every illustration in the British Library's 19th-century book images dataset now has an instance mask: 1.02M clean cutouts of public-domain engravings, no image generation involved! One small segmentation model, $3.24 of compute. Model, masks and a search Space with cutout view all open. 111425
Daniel van Strien @danielvanstrien.bsky.social · 18/08/2026Synthetic data at scale without owning a GPU: datatrove's new Jobs backend + Qwen3.8-27B → 35,837 length-controllable TL;DRs of Hugging Face cards, $0.43 per 1,000. Full guide: danielvanstrien.xyz/posts/2026/d...danielvanstrien.xyzDistilling Qwen3.8 with datatrove on Hugging Face Jobs – Daniel van Striendatatrove’s new Jobs backend plus a days-old 27B teacher: regenerating a 35,837-summary training dataset in one afternoon for $15.58, with a calibration-first workflow and length-controllable outputs. 051
Daniel van Strien @danielvanstrien.bsky.social · 10/08/2026For this project, we want to potentially scale to a very large number of pages and also want the approach to be somewhat reproducible, so we're not really considering closed models. My guess is Gemini would be fairly highly ranking on this leaderboard. 010
Daniel van Strien @danielvanstrien.bsky.social · 10/08/2026A new @hf.co and @eleutherai.bsky.social project, with @storytracer.com! Blog post: huggingface.co/blog/fineboo...huggingface.coFineBooks: are open OCR models good enough to unlock historical knowledge?A Blog post by FineBooks on Hugging Face 142
Daniel van Strien @danielvanstrien.bsky.social · 10/08/2026So we built a leaderboard. 14 open models, scored on 2,165 pages of expert human transcription. The best read historical print at ~97.6% character accuracy, for under $2 per 1,000 pages. 150
Daniel van Strien @danielvanstrien.bsky.social · 10/08/2026FineBooks wants to help re-OCR the public domain to unlock a rich source of data for AI, for researchers, and for enjoyment. Step one: work out which modern OCR models are actually good enough. 2255
Reposted by Daniel van StrienSebastian Majstorovic @storytracer.com · 10/08/2026Are open OCR models good enough to unlock historical knowledge? @danielvanstrien.bsky.social and I built a leaderboard using six expert-transcribed volumes from the @biodivlibrary.bsky.social sky.social. Meet FineBooks, from @hf.co and @eleutherai.bsky.social! huggingface.co/blog/fineboo...huggingface.coFineBooks: are open OCR models good enough to unlock historical knowledge?A Blog post by FineBooks on Hugging Face 06322
Daniel van Strien @danielvanstrien.bsky.social · 07/08/2026Uploaded a dataset of 1,080,814 public domain images, mostly from 19th-century books, to the Hugging Face Hub: huggingface.co/datasets/big... You can also do semantic search against the images here: huggingface.co/spaces/davan... 15412
Daniel van Strien @danielvanstrien.bsky.social · 03/08/2026Shares are relative, so a falling share doesn't mean falling usage; the attributed base is growing. 23% of agent traffic still comes from tools that haven't registered, down from 37% in June. Curious whether the open harnesses close the gap in August! 040
Daniel van Strien @danielvanstrien.bsky.social · 03/08/2026Public data on which coding agents actually use the Hugging Face Hub, refreshed monthly. Claude Code now sends a majority of everything the Hub can attribute to a named agent. In May, it led on 4 days out of 30. In July, all 30. huggingface.co/datasets/hug... 4132
Daniel van Strien @danielvanstrien.bsky.social · 31/07/2026The model describes what's on screen; it can't tell you who anyone is or why a film was made. That part stays human. ~$10 in compute on @hf.co Jobs Unreviewed output: good for finding, not citing. huggingface.co/spaces/davan...huggingface.coPrelinger moments - a Hugging Face Space by davanstrienTimestamped search over VLM-captioned Prelinger films 191
Daniel van Strien @danielvanstrien.bsky.social · 31/07/2026A film catalogue tells you what a film is about, not what happens inside it. So: 1,864 public-domain Prelinger films, described every ~60 seconds by an open 2B video model. 23,148 searchable moments. Search "typing on a computer keyboard", land on the second it happens. 2144
Daniel van Strien @danielvanstrien.bsky.social · 30/07/2026New recipe: timestamped video captions on @hf.co Jobs. Point it at a bucket of videos → parquet dataset out: scene descriptions + second-precise <start – end> events. ~$0.05 per hour of footage on a single A10G (Marlin-2B on vLLM). huggingface.co/datasets/uv-... 091
Reposted by Daniel van StrienDaniel van Strien @danielvanstrien.bsky.social · 14/07/2026Small open models are getting genuinely good at document parsing: OvisOCR2 (0.9B, Apache 2.0) is claiming SOTA on OmniDocBench v1.6. Day-1 recipe: OCR a whole HF image dataset — digitised newspapers, archives, zines — to markdown with one command. huggingface.co/datasets/uv-scripts/ocr 0363
Daniel van Strien @danielvanstrien.bsky.social · 14/07/2026Small open models are getting genuinely good at document parsing: OvisOCR2 (0.9B, Apache 2.0) is claiming SOTA on OmniDocBench v1.6. Day-1 recipe: OCR a whole HF image dataset — digitised newspapers, archives, zines — to markdown with one command. huggingface.co/datasets/uv-scripts/ocr 0363
Daniel van Strien @danielvanstrien.bsky.social · 11/07/2026This model does it e2e so nothing for me to choose! 040
Reposted by Daniel van StrienJim Clifford @jimclifford.bsky.social · 10/07/2026Reading the Archive by Machine: An OCR Benchmark for Historians, 1612–1921 Here is version 1 of a working paper on the new OCR tools that are transforming digital history. working-papers-in-critical-search.github.io/paper-004-oc...working-papers-in-critical-search.github.ioReading the Archive by Machine – Working Papers in Critical SearchA benchmark of six OCR systems (Tesseract, olmOCR 2, Chandra 2, Infinity Parser 2, GLM-OCR, and Gemini 3.5 Flash) on human-transcribed archival documents spanning 1612–1921: early-modern print, ninete... 24114
Reposted by Daniel van StrienJames Feigenbaum @jamesfeigenbaum.bsky.social · 09/07/2026I don't have it yet, but there's an econ history paper in these transcripts or transcripts like them (that we can now turn into textual data at the cost of pennies per hour) 1133
Daniel van Strien @danielvanstrien.bsky.social · 09/07/2026It's unedited machine output over scratchy radio (it keeps hearing "Follow eleven, this is Houston"). The recipe runs on any audio collection in one command: huggingface.co/datasets/uv-...huggingface.couv-scripts/transcription · Datasets at Hugging FaceWe’re on a journey to advance and democratize artificial intelligence through open source and open science. 020
Daniel van Strien @danielvanstrien.bsky.social · 09/07/2026The transcript is a finding aid, not a replacement — every segment links back to the Internet Archive originals. And the pipeline flagged which reels are blank carrier hiss: useful collection metadata in its own right. Dataset (CC0): huggingface.co/datasets/dav...huggingface.codavanstrien/apollo-11-diarized · Datasets at Hugging FaceWe’re on a journey to advance and democratize artificial intelligence through open source and open science. 110
Daniel van Strien @danielvanstrien.bsky.social · 09/07/2026Open ASR models with speaker diarization are now fast and cheap: I diarized 174 hours of Apollo 11 mission audio (the real July 1969 NASA tapes) for $9.46 with a 0.9B open model. Search it, hear any moment on the original tape: huggingface.co/spaces/davan... 3251
Daniel van Strien @danielvanstrien.bsky.social · 07/07/2026Write-up + plus how I ran the whole thing on HF Jobs, no local GPU: danielvanstrien.xyz/posts/2026/o...danielvanstrien.xyzBenchmarking OCR fairly is harder than it looks – Daniel van StrienI ran ten current OCR models on olmOCR-bench’s ‘old scans’ subset. The ranking flips depending on one thing: whether you want a model that keeps what’s on the page, or one that cleans it up. A note on... 072
Daniel van Strien @danielvanstrien.bsky.social · 07/07/2026Two other things: a 1B model (LightOnOCR-2) has the best raw transcription in the field, and PaddleOCR-VL 1.6 sometimes hallucinates Chinese characters on English scans. 130
Daniel van Strien @danielvanstrien.bsky.social · 07/07/2026IMO this is because a lot of VLM-based OCR models were made to provide tokens for training. It's less useful if you want faithful OCR of the whole page, like an archive where the letterhead is part of the record. 220
Daniel van Strien @danielvanstrien.bsky.social · 07/07/2026The score rewards dropping boilerplate, i.e. letterheads, stamps, page numbers, so a model that reads the page more faithfully can rank lower. 120
Daniel van Strien @danielvanstrien.bsky.social · 07/07/2026On the headline score, PaddleOCR-VL beats NuExtract3 (38.6 vs 37.8). But rank by how much of the page each model actually reads, and NuExtract3 is well ahead (41.6 vs 31.2). Same two models, opposite order. 110
Daniel van Strien @danielvanstrien.bsky.social · 07/07/2026I ran 10 newer OCR models on @ai2.bsky.social's olmOCR-bench "old scans" subset. The ranking flips depending on what you actually want. 3193
Daniel van Strien @danielvanstrien.bsky.social · 02/07/2026Explore it in the Dataset Viewer, no code needed: - the monthly leaderboard + how it shifts as new tools launch - request share vs user share (a few heavy pipelines vs many light users) - daily data: launch spikes, weekday vs weekend patterns... huggingface.co/datasets/hug...huggingface.cohuggingface/agent-usage · Datasets at Hugging FaceWe’re on a journey to advance and democratize artificial intelligence through open source and open science. 010
Daniel van Strien @danielvanstrien.bsky.social · 02/07/2026Coding agents are real users of the @hf.co Hub! They're searching for models, building and pushing datasets, training models on Jobs, spinning up Spaces... Now there's public data: each agent's share of Hub traffic, updated monthly 👇 2201
Daniel van Strien @danielvanstrien.bsky.social · 02/07/2026Can run it locally. Just streams the data from HF so you don't need the space to store common crawl locally! 110
Daniel van Strien @danielvanstrien.bsky.social · 02/07/2026The latest @commoncrawl.bsky.social crawl indexes way more than HTML pages. 20.9M PDFs. Plus calendars, BibTeX, markdown... The whole index now lives in a @hf.co Bucket, so I pulled this with one SQL query straight over it (new S3 API). 2.1B rows, nothing downloaded, $0 to read. 2246
Daniel van Strien @danielvanstrien.bsky.social · 30/06/2026You can now use 100s of tools with @hf.co Buckets, thanks to the new S3 API! Usually just one or two lines to change. huggingface.co/docs/hub/sto... 060
Daniel van Strien @danielvanstrien.bsky.social · 29/06/2026A model that reads the page more faithfully can rank lower. Works well as a score if you want clean "reading-order" tokens for LLM training, but less useful as a metric if you care about is "faithful" OCR. 010
