> Mistral OCR 3 is ideal for both high-volume enterprise pipelines and interactive document workflows.
I don’t know how they can make this statement with 79% accuracy rate. For any serious use case, this is an unacceptable number.
I work with scientific journals and issues like 2.9+0.5 and 29+0.5 is something we regularly run into that has us never being able to fully trust automated processes and require human verification every step.
And I believe the number is 74%, compared to OCR 2.
What matters is whether this is better than competition/alternatives. Of course nobody is just going to take the output as is. If you do that, that's your problem.
I've worked on document extraction a lot and while the tweet is too flippant for my taste, it's not wrong. Mistral is comparing itself to non-VLM computer vision services. While not necessarily what everyone needs, they are a very different beasts compared to VLM based extraction because it gives you precise bounding boxes, usually at the cost of larger "document understanding".
Its failure mode are also vastly different. VLM-based extraction can misread entire sentences or miss entire paragraphs. Sonnet 3 had that issue. Computer vision models instead will make in-word typos.
there has been so many open source OCR in the last 3 months that would be good to compare to those especially when some are not even 1B params and can be run on edge devices.
- paddleOCR-VL
- olmOCR-2
- chandra
- dots.ocr
I kind of miss there is not many leaderboard sections or arena for OCR and CV and providers hosting those. Neglected on both Artificial Analysis and OpenRouter.
what I like in MistralOCR is that they have simple pricing $1/1k pages and API hosted on their servers. With other OCR is hard to compare pricing because are token based and you don't know how many tokens is the image unless you run your own test.
E.g. with Gemini 3.0 flash you might seem that model pricing increased only slightly comparing to Gemini 2.5 flash until you test it and will see that what used to be 258 per 384x384 input tokens now is around 3x more.
I spent like three hours trying to get one of these running and then gave up. I think the paddleOCR one.
It took an hour and a half to install 12 gigabytes of pytorch dependencies that can't even run on my device, and then it told me it had some sort of versioning conflict. (I think I was supposed to use UV, but I had run out of steam by that point.)
Maybe I should have asked Claude to install it for me. I gave Claude root on a $3 VPS, and it seems to enjoy the sysadmin stuff a lot more than I do...
Incidentally I had a similar experience installing open web UI... It installed 12 GB of pytorch crap.. I rage quit and deleted the whole thing, and replicated the functionality I actually needed in 100 lines of HTML.... Too bad I can't do that with OCR ;)
It seems like Mistral is just chasing around sort of "the fringes" of what could be useful AI features. Are they just getting out-classed by OAI, Google, Anthropic?
It seems like EU in general should be heavily invested in Mistral's development, but it doesn't seem like they are.
Yep. I saw the title and got excited.... this is a particular problem area where I think these things can be very effective. There are so many data entry class tasks which don't require huge knowledge or judgement... just clear parsing and putting that into a more machine digestible form.
I don't know... feels like this sort of area, while not nearly so sexy as video production or coding or (etc.)... but seems like reaching a better-than-human performance level should be easier for these kinds of workloads.
Following the leaders too closely seems like a bad move, at least until a profitable business model for an AI model training company is discovered. Mistral’s models are pretty good, right? I mean they don’t have all the scaffolding around them that something like chatGPT does, but building all that scaffolding could be wasted effort until a profitable business model is shown.
Until then, they seem to be able to keep enough talent in the EU to train reasonably good models. The kernel is there, which seems like the attainable goal.
I think there is a lot of broad support, but they're just kind of hamstrung by EU regulation on AI development at this stage. I think the end game will ultimately be getting acquired by an American company, and then relocating.
Is open router still sending all OCR jobs to Mistral? I wonder if they're trying to keep that spot. Seems like Mistral and Google are the best at OCR right now, with Google leading Mistral by a fair bit.
Gave it a birth registry from a Portuguese locality from 1755 which my dad and I often decipher to figure out geneology and it did a terrible job.
Regular Gemini Thinking can actually get 70-80% of the documents correct except lots of mistakes on given names. Chatgpt maybe understands like 50-60%.
This Mistral model butchered the whole text, literally not a word was usable. To the point I think I'm doing something wrong.
It's tough but my dad is quite good at it. He has books of common abbreviations and agglutinations from different centuries. After you get used to it it's faster and very fun.
I am too. Gemini 3.0 fast on old scrawled diary entries in English from 100+ years ago got them 95% right. It also added historical context when I prefaced the images with the identity of the writer, such as summaries of an old military unit history in Europe post-WW1 it got from a very obscure U.S. Army archive.
I don’t know how they can make this statement with 79% accuracy rate. For any serious use case, this is an unacceptable number.
I work with scientific journals and issues like 2.9+0.5 and 29+0.5 is something we regularly run into that has us never being able to fully trust automated processes and require human verification every step.
What matters is whether this is better than competition/alternatives. Of course nobody is just going to take the output as is. If you do that, that's your problem.
> can someone help folks at Mistral find more weak baselines to add here? since they can't stomach comparing with SoTA....
> (in case y'all wanna fix it: Chandra, dots.ocr, olmOCR, MinerU, Monkey OCR, and PaddleOCR are a good start)
Its failure mode are also vastly different. VLM-based extraction can misread entire sentences or miss entire paragraphs. Sonnet 3 had that issue. Computer vision models instead will make in-word typos.
- paddleOCR-VL
- olmOCR-2
- chandra
- dots.ocr
I kind of miss there is not many leaderboard sections or arena for OCR and CV and providers hosting those. Neglected on both Artificial Analysis and OpenRouter.
https://www.ocrarena.ai/leaderboard
Hasn't been updated for Mistral but so far gemeni seems to top the leaderboard.
E.g. with Gemini 3.0 flash you might seem that model pricing increased only slightly comparing to Gemini 2.5 flash until you test it and will see that what used to be 258 per 384x384 input tokens now is around 3x more.
It took an hour and a half to install 12 gigabytes of pytorch dependencies that can't even run on my device, and then it told me it had some sort of versioning conflict. (I think I was supposed to use UV, but I had run out of steam by that point.)
Maybe I should have asked Claude to install it for me. I gave Claude root on a $3 VPS, and it seems to enjoy the sysadmin stuff a lot more than I do...
Incidentally I had a similar experience installing open web UI... It installed 12 GB of pytorch crap.. I rage quit and deleted the whole thing, and replicated the functionality I actually needed in 100 lines of HTML.... Too bad I can't do that with OCR ;)
https://www.codesota.com/ocr
It seems like EU in general should be heavily invested in Mistral's development, but it doesn't seem like they are.
I don't know... feels like this sort of area, while not nearly so sexy as video production or coding or (etc.)... but seems like reaching a better-than-human performance level should be easier for these kinds of workloads.
Until then, they seem to be able to keep enough talent in the EU to train reasonably good models. The kernel is there, which seems like the attainable goal.
Are they? IIRC their best model is still worse than the gpt-oss-120B?
Maybe, i think it will be to our benefit when the bubble pops that we are not heavily invested, no harm investing a little.
- table entries hallucinated - tables messed up (tables merged, forgot rows) - forgot to parse some text passages
if you are doing something serious, i would not use it
1. Use native PDF parsing if the model supports it
2. Use this Mistral OCR model (we updated to this version yesterday)
3. UNLESS you override the "engine" param to use an alternate. We support a JS-based (non-LLM) parser as well [0]
So yes, in practice a lot of OCR jobs go to Mistral, but not all of them.
Would love to hear requests for other parsers if folks have them!
[0] https://openrouter.ai/docs/guides/overview/multimodal/pdfs#p...
Regular Gemini Thinking can actually get 70-80% of the documents correct except lots of mistakes on given names. Chatgpt maybe understands like 50-60%.
This Mistral model butchered the whole text, literally not a word was usable. To the point I think I'm doing something wrong.
The test document: https://files.fm/u/3hduyg65a5
We were mind blown how good Gemini was at it.
Huge timesaver.