Show HN: OCR Arena – A playground for OCR models • Hacker News Board

Show HN: OCR Arena – A playground for OCR models

#1

Post by kbyatnal » Fri, Nov 21, 2025, 4:44 PM UTC

I built OCR Arena as a free playground for the community to compare leading foundation VLMs and open-source OCR models side-by-side.

Upload any doc, measure accuracy, and (optionally) vote for the models on a public leaderboard.

It currently has Gemini 3, dots.ocr, DeepSeek, GPT5, olmOCR 2, Qwen, and a few others. If there's any others you'd like included, let me know!

Show HN: OCR Arena – A playground for OCR models
ocrarena.ai

Re: Show HN: OCR Arena – A playground for OCR models

#2

Post by ylhert » Fri, Nov 21, 2025, 5:04 PM UTC

dang wrote: post #5

[under-the-rug stub] [see https://news.ycombinator.com/item?id=45988611 for explanation]

We've got like 10 LLM arenas but nothing for OCR yet, really hope this takes off!

Re: Show HN: OCR Arena – A playground for OCR models

#3

Post by profburial » Fri, Nov 21, 2025, 6:05 PM UTC

dang wrote: post #5

[under-the-rug stub] [see https://news.ycombinator.com/item?id=45988611 for explanation]

This is a killer idea!

Re: Show HN: OCR Arena – A playground for OCR models

#4

Post by athoscouto » Fri, Nov 21, 2025, 10:58 PM UTC

dang wrote: post #5

[under-the-rug stub] [see https://news.ycombinator.com/item?id=45988611 for explanation]

Nice! Would love to see Azure Document Intelligence on this

Re: Show HN: OCR Arena – A playground for OCR models

#5

Post by dang » Mon, Nov 24, 2025, 10:37 PM UTC

[under-the-rug stub]

[see https://news.ycombinator.com/item?id=45988611 for explanation]

Re: Show HN: OCR Arena – A playground for OCR models

#6

Post by arathis » Mon, Nov 24, 2025, 10:52 PM UTC

Claude would be good!

Re: Show HN: OCR Arena – A playground for OCR models

#7

Post by ianhawes » Mon, Nov 24, 2025, 11:25 PM UTC

Please add Chandra by Datalab

Re: Show HN: OCR Arena – A playground for OCR models

#8

Post by fzysingularity » Mon, Nov 24, 2025, 11:25 PM UTC

FYI one of the models on the battle was pretty slow to load. Are these also being rated on latency or just quality?

Re: Show HN: OCR Arena – A playground for OCR models

#9

Post by zzleeper » Mon, Nov 24, 2025, 11:39 PM UTC

Love this! Would have liked to see something like textract for a pre-LLM benchmark (but of course that's expensive), and also a distinction between handwritten text and printed one.

But still, this is incredibly useful!

Re: Show HN: OCR Arena – A playground for OCR models

#10

Post by krashidov » Mon, Nov 24, 2025, 11:43 PM UTC

I would be curious to see how Sonnet does. Their models are pretty solid when it comes to PDFs