Birbla

Join the discussion

Write your take first — we'll ask for email only when you're ready to publish.

Sign in
  • Hacker News
  • Is this similar to CleanshotX?
    by jbverschoor
  • Love how this is just a chrome extension... Just made an API so that you could easily OCR everything with SOTA results (using finetuned VLM) at 1/3rd of the usual costs... Would absolutely love to chat and see if we can help out !
    by maestroquirk
  • https://learn.microsoft.com/en-us/windows/powertoys/text-ext...
    by shimonabi
  • Hi guys, the chrome extension was finally approved and is now available at https://chromewebstore.google.com/detail/ocr-it/pekiamimgjbj...

    We are still waiting for the firefox extension to be approved, i will post it here whenever we hear something. In the meanwhile it can be tested with the zip file here https://github.com/thiagotigaz/ocr-it/releases or by building manually.

    by thiagolima
  • Does anyone have suggestions on how I could OCR lots of handwritten math notes with diagrams? I have tons of PDFs waiting for me to manually type them myself and can't justify dedicating weeks to do it.
    by kalinkochnev
  • Most LLMs can do a pretty good job of this out of the box.

    If you have a coding agent available, ask it to try transcribing a few of the PDFs.

    by danbrooks
  • I've seen papers using fine-tuned small local vision models to transcribe math into latex. Personally I tried Claude Opus & Claude Code directly to do this. It works (to my surprise) but AI sometimes take guesses that are mathematically right but deviates with my original writing. There is no 100% correct method though.
    by xiaoyu2006
  • you could try using a local vision model, like Mage-VL from microsoft. Its only a 5b model so its quite small for the capability it has.
    by nshotton
  • I tested many open-source and hosted OCR models and Datalab Chandra was the most accurate. It can parse complex layouts, tables, handwriting, and formulas at a fraction of the cost of Claude/Gemini.

    Local: https://github.com/datalab-to/chandra Hosted: https://www.datalab.to

    Another decent option is GLM OCR. It's slightly less accurate but faster and cheaper.

    Local: https://github.com/zai-org/GLM-OCR Hosted: https://docs.z.ai/guides/vlm/glm-ocr

    Other models such as PaddleOCR, dots.ocr and DeepSeek OCR performed significantly worse.

    by phenomen
  • I added Firefox support. 0.3.0 builds for both browsers from the same source, and it's submitted for both google/chrome and mozilla/firefox, i am waiting on reviews now, which usually takes a few days.

    Until it's approved you guys can download the ready to use releases:

    Download ocr-it-firefox-0.3.0.zip from https://github.com/thiagotigaz/ocr-it/releases/tag/v0.3.0

    If you'd rather build from source, the steps are in the README: https://github.com/thiagotigaz/ocr-it#install

    by thiagolima
  • Great news!
    by HackerThemAll
  • If you use a Linux desktop (I am on Fedora), Gradia[1][2] is definitely worth a look as well.

    It has a similar workflow for taking screenshots and then immediately annotating or editing them, without having to open a separate image editor. And: it provides also an local OCR feature (which is why I comment this here), you can extract text from a screenshot with on-screen OCR using Tesseract with the small button beside the "Crop Image" one.

    Combined with the syntax-highlighting feature for screenshots of code snippets, the OCR is surprisingly useful in combination if you e.g. quickly discuss some code in a chat when copy is blocked for whatever reason (e.g. somone sent you a screenshot in the first place).

    [1] https://gradia.alexandervanhee.be/

    [2] https://flathub.org/en/apps/be.alexandervanhee.gradia

    Edit: fixed wrong link index numbers

    by andreashaerter

  •   “Pin a region once. Hit a hotkey on every page. Get the whole book as text.”
    Much better than the old definition of “region lock”, nice.

    HN isn’t a fan of the generated readmes though, though vibed software (thoroughly used) can be all good.

    by Barbing
  • it can also auto paginate for you, no need to keep hitting the hotkey every page. It can paginate by hotkey, xy point on screen or selector.
    by thiagolima
  • > though vibed software (thoroughly used) can be all good.

    Yes, but the problem with these vibe-coded crap is that they are pretty much always less than a week old, which means it wasn't even used before the “author” submitted it here.

    (The author didn't even bother writing their comment themselves by the way: https://news.ycombinator.com/item?id=49415857)

    by stymaar
  • Also available natively to the OS (Windows) with PowerToys, if you want an alternative to a browser extension. One of the unsung heroes of that library.

    Jury is still out on which is more trustworthy handling any personal data, Microsoft or Google. Neither.

    by tobinfekkes
  • And Plasma Spectacle does it too
    by agilob
  • Is Tesseract still the best choice for local OCR in 2026? I was always underwhelmed with its real-world performance.
    by rickcarlino
  • PaddlePaddle (v6) is fantastic and fast
    by deivid
  • It works very well for me. It handles non-English characters and diacritics without issues.
    by HackerThemAll
  • We did some benchmarking against other libraries like RapidOCR and EasyOCR results in https://thiagotigaz.github.io/ocr-it/bench/
    by thiagolima
  • There's EasyOCR and RapidOCR too, I guess benchmark and see what's best for your material? Oh and Multimodal LLMs :)
    by _joel
  • Definitely not. Even Chrome has a built in OCR that performs amazingly. I got an LLM to write a quick python wrapper to it [1], so I'm sure you should be able to access it from an extension

    [1] https://github.com/sergiocorreia/clv-locro

    by zzleeper
  • I, at this point, use Qwen2.5-VL-3B-Instruct for most of the small OCR I want to do. It is much much better than my experience with Tesseract in general. The nice thing about it is that if you give it, say, a movie poster you can ask for the "title of the movie" and it will, to the best of its ability, do just that, no need for regex or filtering after. For smallish images after loading the 3B model runs in <1 second. 7B takes longer but is obviously more accurate.

    I might be a bit behind, all of this is from early this year for the most part, but for something like "I have 3000 movie posters and I want to get the titles with like 90% accuracy" it is good (much better than Tesseract), and it'll do that in like an hour.

    EDIT: I guess one thing is Tesseract will kind of give gibberish back when it fails. The main issue with the LLMs are that instead they take a stab at it (like for a movie poster it'll give part of a quote, or a actor name) back. Makes knowing when it fails a little harder. As long as you have some way to verify when it is likely failing they are very good though.

    by LeonardoTolstoy