←back to thread

DeepSeek OCR

(github.com)
990 points pierre | 7 comments | 20 Oct 25 06:26 UTC | HN request time: 0.529s | source | bottom
Show context
pietz ◴[20 Oct 25 08:49 UTC] No.45641449[source]▶
My impression is that OCR is basically solved at this point.

The OmniAI benchmark that's also referenced here wasn't updated with new models since February 2025. I assume that's because general purpose LLMs have gotten better at OCR than their own OCR product.

I've been able to solve a broad range of OCR tasks by simply sending each page as an image to Gemini 2.5 Flash Lite and asking it nicely to extract the content in Markdown under some additional formatting instructions. That will cost you around $0.20 for 1000 pages in batch mode and the results have been great.

I'd be interested to hear where OCR still struggles today.

replies(23): >>45641470 #>>45641479 #>>45641533 #>>45641536 #>>45641612 #>>45641806 #>>45641890 #>>45641904 #>>45642270 #>>45642699 #>>45642756 #>>45643016 #>>45643911 #>>45643964 #>>45644404 #>>45644848 #>>45645032 #>>45645325 #>>45646756 #>>45647189 #>>45647776 #>>45650079 #>>45651460 #
kbumsik ◴[20 Oct 25 08:51 UTC] No.45641470[source]▶
> My impression is that OCR is basically solved at this point.

Not really in practice to me. Especially they still struggle with Table format detection.

replies(2): >>45641501 #>>45643548 #
coulix ◴[20 Oct 25 08:54 UTC] No.45641501[source]▶
This.

Any complex parent table span cell relationship still has low accuracy.

Try the reverse, take a complex picture table and ask Chatgpt5, claude Opus 3.1, Gemini Pro 2.5 to produce a HTML table.

They will fail.

replies(2): >>45641541 #>>45641916 #
1. pietz ◴[20 Oct 25 09:40 UTC] No.45641916[source]▶
Maybe my imagination is limited or our documents aren't complex enough, but are we talking about realistic written documents? I'm sure you can take a screenshot of a very complex spreadsheet and it fails, but in that case you already have the data in structured form anyway, no?
replies(2): >>45642356 #>>45644170 #
2. kbumsik ◴[20 Oct 25 10:39 UTC] No.45642356[source]▶
> realistic written documents?

Just get a DEF 14A (Annual meeting) filing of a company from SEC EDGAR.

I have seen so many mistakes when looking at the result closely.

Here is a DEF 14A filing from Salseforce. You can print it to a PDF and then try converting.

https://www.sec.gov/Archives/edgar/data/1108524/000110852425...

replies(1): >>45643178 #
3. grosswait ◴[20 Oct 25 12:34 UTC] No.45643178[source]▶
Historical filings are still a problem, but hasn’t the SEC required filing in an XML format since the end of 2024?
replies(1): >>45643659 #
4. richardlblair ◴[20 Oct 25 13:28 UTC] No.45643659{3}[source]▶
It's not really about SEC filings, though. While we folks on HN would never think of hard copies of invoices, but much of the world still operates this way.

As mentioned above I have about 200 construction invoices. They are all formatted in a way that doesn't make sense. Most fail both OCR and OpenAI

replies(1): >>45645579 #
5. daemonologist ◴[20 Oct 25 14:17 UTC] No.45644170[source]▶
Now if someone mails or faxes you that spreadsheet? You're screwed.

Spreadsheets are not the biggest problem though, as they have a reliable 2-dimensional grid - at worst some cells will be combined. The form layouts and n-dimensional table structures you can find on medical and insurance documents are truly unhinged. I've seen documents that I struggled to interpret.

replies(1): >>45645458 #
6. KoolKat23 ◴[20 Oct 25 16:03 UTC] No.45645458[source]▶
To be fair, this is problematic for humans too. My old insurer outright rejected things like that stating it's not legible.

(I imagine it also had the benefit of reducing fraud/errors).

In this day and age, it's probably easier/better to change the process around that as there's little excuse for such shit quality input. I understand this isn't always possible though.

7. KoolKat23 ◴[20 Oct 25 16:13 UTC] No.45645579{4}[source]▶
OpenAI has unusuably low image DPI. Try Gemini.