@danielvanstrien.bsky.social - Daniel van Strien

07 Sep 2026 15:43 +0000

OCR 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: https://huggingface.co/collections/davanstrien/ocr-on-the-hub

03 Sep 2026 14:53 +0000

A 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. https://huggingface.co/spaces/finebooks/bhl-ocr-leaderboard

26 Aug 2026 19:19 +0000

Every 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. [contains quote post or other embedded content]

25 Aug 2026 17:01 +0000

Uploaded a dataset of 115,293 illustrated pages from the Encyclopaedia Britannica, 1st edition (1768) to 14th (1929) to the Hub https://huggingface.co/datasets/biglam/britannica-illustrated-pages

20 Aug 2026 10:42 +0000

Every 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.

18 Aug 2026 08:24 +0000

Synthetic 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: https://danielvanstrien.xyz/posts/2026/distilling-qwen38-datatrove-jobs/

10 Aug 2026 12:18 +0000

FineBooks 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.

07 Aug 2026 17:18 +0000

Uploaded a dataset of 1,080,814 public domain images, mostly from 19th-century books, to the Hugging Face Hub: https://huggingface.co/datasets/biglam/british-library-book-images You can also do semantic search against the images here: https://huggingface.co/spaces/davanstrien/bl-images-search

03 Aug 2026 11:48 +0000

Public 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. https://huggingface.co/datasets/huggingface/agent-usage

31 Jul 2026 11:44 +0000

A 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.

30 Jul 2026 08:31 +0000

New recipe: timestamped video captions on @hf.co Jobs. Point it at a bucket of videos → parquet dataset out: scene descriptions + second-precise events. ~$0.05 per hour of footage on a single A10G (Marlin-2B on vLLM). https://huggingface.co/datasets/uv-scripts/video

14 Jul 2026 15:24 +0000

Small 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

14 Jul 2026 15:24 +0000

09 Jul 2026 16:41 +0000

Open 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: https://huggingface.co/spaces/davanstrien/apollo-11-search

07 Jul 2026 16:36 +0000

I ran 10 newer OCR models on @ai2.bsky.social's olmOCR-bench "old scans" subset. The ranking flips depending on what you actually want.

02 Jul 2026 15:37 +0000

Coding 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 👇

02 Jul 2026 07:19 +0000

The 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.

30 Jun 2026 12:18 +0000

You can now use 100s of tools with @hf.co Buckets, thanks to the new S3 API! Usually just one or two lines to change. https://huggingface.co/docs/hub/storage-buckets-s3

29 Jun 2026 14:11 +0000

Fairly benchmarking OCR models is hard! Ran a few newer OCR models on @ai2.bsky.social's olmOCR-bench "old scans" subset. Worth knowing: the score "punishes" models for extracting too much (letterheads, stamps, etc.)

24 Jun 2026 16:24 +0000

If libraries, archives and museums pooled their (labelled) data, they could build state-of-the-art open models for the things they actually care about! I tried a small version: one open model (NuExtract-3, 4B) fine-tuned to read archival index cards across several collections.

24 Jun 2026 16:24 +0000

23 Jun 2026 08:36 +0000

Ran the same 650M model across 7 languages: German, Polish, Finnish, Russian, Serbian, Estonian, Swedish via @europeana.bsky.social newspapers. Still some challenging layouts, but pretty mind-blowing to me that a 650M model is doing this when a year ago 72B VLMs failed very often. [contains quote post or other embedded content]

22 Jun 2026 15:16 +0000

I think VLM-based OCR might finally be close to working on historic newspapers! Many models I've tried before failed i.e. hallucinations, repetition loops, context overflow. Surya OCR 2 (a 650M model!) does a very good job!