Rendered at 04:30:54 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
ComputerPerson 11 hours ago [-]
I've got a scan from a book that I OCR with new releases. Ligatures, critical sigla, Fraktur letterforms, subscripts, superscripts, etc.
Nothing special about this model for overly-detailed work like mine.
It's been a while since I last tested (and discontinued my subscription), but the "pro" models from OpenAI dominate. Not surprising, given the price difference, but it would be nice if an OCR-specific model could perform better. It's worth mentioning that even the highest-end models do a pretty poor job with intricate text like mine.
SyneRyder 8 hours ago [-]
While I haven't tried OpenAI for OCR, I've put my small scale OCR work through both Claude and Mistral OCR. Claude is absolutely better - even in OCR work I did last week and compared with Mistral OCR 4.0.
Mistral's one advantage is that Anthropic now flags OCR, because they don't allow anything that could be considered "reproduction", even of work for which you own the copyright. So my new workflow is Mistral OCR for the actual OCR, followed by a proofreading pass by Claude (which is allowed). Claude is obviously more expensive, but it caught entirely hallucinated sentences created by Mistral OCR 4.0, so I was glad for the backup check.
Oras 8 hours ago [-]
Same company that OCRed millions of books, the irony.
I feel Anthropic is destroying itself with all these restriction. They got away because their models were the best for coding, but that is not an advantage anymore as OpenAI and other open source are already better.
usef- 3 hours ago [-]
They were criticised/sued early on when people could reproduce copyright things. I dont think in this case it's something they'd prefer to do?
SyneRyder 17 minutes ago [-]
Yep, my understanding is that many guardrails like this are actually the result of government legislation (eg the Fable bans) or terms of settling copyright lawsuits over reproducing copyrighted text and lyrics.
I mentioned it in a sibling reply, but here's Anthropic's support document about not using Claude to reproduce content verbatim that already exists, regardless of copyright.
> Claude is obviously more expensive, but it caught entirely hallucinated sentences created by Mistral OCR 4.0, so I was glad for the backup check.
What does this entail? What does Claude do to decide that the text it was provided was hallucinated? Are you telling Claude that the source was OCR'd by another LLM?
SyneRyder 24 minutes ago [-]
I'm basically doing the OCR twice, except in the Claude proofreading pass, it is not being asked to transcribe the document to a Markdown file. I'm pointing it to the same image input files, and to the Mistral OCR transcript Markdown file (it knows it's a Mistral OCR output), and ask Claude to check that the text is correct and point out the errors - and then make the necessary edits.
I can't speak for Mistral OCR 4.1, but the hallucinations in 4.0 were so egregious (just completely making up new sentences in the middle of a page) that I knew I can't trust Mistral OCR on its own.
mattnewton 5 hours ago [-]
I assume they pass the image alongside the text to redo/recheck the work. Yes this is silly, but is apparently required to get around refusals.
jassyr 6 hours ago [-]
>Anthropic now flags OCR
I haven't seen any difference in my ocr workflows, what do you mean by this?
tkgally 1 hours ago [-]
Not the person you’re responding to, but I’ve had Claude refuse to OCR pages from in-copyright books. I was sometimes (but not always) able to get around that by changing models, by telling it that I was doing the text conversion only for personal use, or by first telling it to use Tesseract or another OCR engine to do the initial pass and then having a Claude subagent proofread and clean up the OCR output.
I’ve also had it refuse to OCR public-domain books that included content that it didn’t like, such as references to prostitution in 19th-century books about Japan.
I had one session where Claude refused to continue after it hit some kind of guiderail restriction. I couldn’t see what the trigger was, so I started a new session, gave Claude the link to the previous session, and asked it to diagnose the problem. This new Claude said it couldn’t view the exact guardrail issue, but it did suggest a workaround that turned out to be effective.
laichzeit0 58 minutes ago [-]
Yeah I do something similar and Claude (well OpenAI too) just refuses to transcribe anything related to slavery and pederasty in Ancient Greece.
SyneRyder 33 minutes ago [-]
The response / conversation gets blocked by the guardrails with "API Error: 400 Output blocked by content filtering policy"
At first I thought it was something in the scanned content that was being flagged, but it was the attempt to transcribe that was itself being flagged. Anthropic mention it on their pages:
"Anthropic takes these steps because Claude’s purpose is to generate new content and ideas, not to reproduce content that already exists."
Side note - Claude itself is not aware of this policy, and is unable to see the API responses - the turn just ends. Which turned into a really bizarre failure state where Claude thought I was gaslighting it and kept insisting it could do the work and even had the entire text in memory. Every time it would go to show me and prove it, it would hit API Error 400. I was only able to convince Claude by showing screenshots of my Claude Code screen output so it could see that I was seeing API errors. I've never seen Claude get into that angry & snarky state before, and I hope it doesn't happen again.
kergonath 9 hours ago [-]
I have been quite happy with Mistral OCR for the documents I needed to process (typeset, but old, with questionable scan quality, sometimes elaborate typesetting or, much worse, typewriter-and-handwriting approximations of it). I do not test every new model when they are released, but I did a review shortly after Mistral OCR 3 was released and it was a very good compromise: cheap, fast, and good results without further processing. I found generalist models to be way too much faf to get them to avoid unnecessary modifications to the text and report accurate bounding boxes for figures and tables.
That said, models have sometimes surprising weaknesses and a model could be terrible overall but magically work for one type of document.
kmitz 10 hours ago [-]
I got the opposite experience very recently : tried to OCR a bunch of handwritten emails addresses with chatGPT and I had to make so many corrections that I gave up. Whereas Mistral nailed it on first pass.
usef- 3 hours ago [-]
Which chatgpt model?
rtaylorgarlock 10 hours ago [-]
Yet: how is pricing?
Evaluating contents and routing appropriately isn't a new challenge in OCR, one of the oldest fields of applications in ML.
Thus, how do the smaller open models perform in tandem with relatively pricy $/pg models & APIs?
Your use case is remarkably rare relative to the volume and price sensitivity of enterprise data warehouse ops.
giancarlostoro 8 hours ago [-]
The big difference is traditional OCR used basic pattern matching to find text, whereas models like Mistral OCR (and GPT, etc) use computer vision instead and deep learning to parse text, math equations, and apparently in some cases extract images too.
I'd love to see some advancements in traditional OCR based on ideas and concepts we've learned from newer "OCR-like" models since traditional OCR is drastically cheaper.
10 hours ago [-]
petcat 11 hours ago [-]
> the "pro" models from OpenAI dominate. Not surprising considering the price difference, but it would ne nice if an OCR-specific model could do better.
I haven't been impressed with any of Mistral's models. They obviously realized that they couldn't compete at the frontier so they decided to go for smaller focused models but even those have not been that good.
booi 8 hours ago [-]
This is what I found as well.
We moved away from Cursor but I was looking for a model that would help with FIM (fill-in-middle) multiline autocompletion and people were recommending Mistral's Codestral. We gave it a shot and it was lackluster at best.. Even Google's Gemini did a significantly better job than Codestral.
Ultimately Opus-class models got good enough and I don't do much manual coding anymore.
razemio 6 hours ago [-]
I made a benchmark for handwriting recognition for a project while keeping line breaks and errors (grammar, spelling). Sonnet absolutely dominates it since a good half a year. 5.6 did not change that for me. This should also translate to better ocr.
ComputerPerson 5 hours ago [-]
You're the second person to mention handwriting. I think it might be handwriting-specific; perhaps Anthropic has a better corpus for this.
I really wouldn't know, though. Anthropic models barf out copyright issues for my use case, so I'm unable even to benchmark them. It's a common problem when you're scanning public domain books. Mine are reference texts often cited.
giancarlostoro 8 hours ago [-]
So is Mistral OCR the best one? Have any other OCR models caught some of what you describe? I've been kind of interested in how "OCR" type models work compared to old school OCR.
ComputerPerson 8 hours ago [-]
My use case isn't in the realm of old-school OCR, so it's not a good comparison, but anyway:
As another user pointed out, it's surprisingly random (task-specific). Llama Scout outperformed Gemini Flash 2.5 on a benchmark I built at the time. I didn't include an OCR models.
Mistral might indeed be the best OCR-specific model for my task, now that you ask. Funny. It's so bad at my work that I didn't register it might be the best in its category. This is just based on vibes from my single scan.
bugglebeetle 8 hours ago [-]
The datalab models are the best ones.
josu 8 hours ago [-]
Whats the best open OCR at the moment?
53 minutes ago [-]
king_crimson 11 hours ago [-]
At this point I lost all hope for Europe playing any significant role in the AI race. If that’s a good or a bad thing I don’t know, but it seems to me like that’s the reality.
hadlock 8 hours ago [-]
Much like spaceflight, aerospace or nuclear engineering, you need to retain local talent for national defense purposes. Being 70% as good is still way better than being 100% dependent and heavily leveraged by your opponents.
dylan604 5 hours ago [-]
Being 70% as good in nuclear engineering sounds scary AF. How would you rank Chernobyl? Better or worse than 70%?
lavieestdure 2 hours ago [-]
Being 70% as good does not mean causing accidents. Accidents are caused by systematic operational failures and fundamental design flaws.
dylan604 2 hours ago [-]
If your employees are only 70% as good, as skilled, as knowledgeable while not knowing it, then you can absolutely kill a reactor just like Chernobyl. The overconfidence is what allowed them to put the reactor in the state they did and not realize the implications.
kubb 11 hours ago [-]
It's not a race. You don't get anything for winning.
InsideOutSanta 8 hours ago [-]
Well, you get hundreds of billions in debt, and then you get open weight models distilling your proprietary models.
BenzeneDream 9 hours ago [-]
? It absolutely is a race. Whether thats a positive thing or not is debatable but every lab is definitely in a race. What prize do you win? Imagine a world where only one country has AGI/ASI. Or a world where Europe only gets access to frontier models 6 months later. Far from ideal.
ux266478 9 hours ago [-]
> What prize do you win? Imagine a world where only one country has AGI/ASI.
"Winning the race" doesn't give you that in any meaningful capacity. It gives you, at the absolute most, a temporary window where that's the case. See: nuclear weapons.
booi 8 hours ago [-]
Citing nuclear weapons isn't the flex you think it is. There are only 9 countries that have nuclear weapons and they absolutely flex this power over non-nuclear powers (see Ukraine, Germany, SE Asia etc..)
Europe already has an innovation problem that's already causing structural economic instabilities which Germany has been struggling (and lately failing) to prop up.
As much as it pains me to say this, AI is already a tech revolution and it seems like Europe is just ignoring it. There's more innovation in 3 blocks in downtown San Francisco than the entire continent of Europe.
ux266478 7 hours ago [-]
> There are only 9 countries that have nuclear weapons
QED being first didn't grant exclusivity.
The premise wasn't that AGI isn't useful. There are actually layers to the metaphor where first-mover advantage of AGI is even less meaningful than it was for nuclear weapons, but I leave those as an exercise to the reader to discover.
fn-mote 7 hours ago [-]
> first-mover advantage of AGI is even less meaningful
This is your opinion. Lots of tech money appears to disagree.
You should at least try to explain your contrarian position, or cite your favorite source that makes an argument that you believe.
ux266478 3 hours ago [-]
> You should at least try to explain your contrarian position
FWIW I explained it fully and provided the argument, along with explicit grounding examples. Can you tell me what exactly you don't understand?
To quote your other post:
> This is the weakest part of the argument. Absolutely no reason to believe the second inventor will be 10x faster.
Nor is there any reason to believe that it will work at any usable speed. An LLM capable of AGI running at 0.000001tok/s isn't a very good head start. Nor is it a good moat if it can't actually realize anything material fast enough. Lord knows America has labor problems.
I'm demonstrating the context is a lot more complicated than "be first". There are innumerable specific examples. They're not guaranteed to happen, that's not the point being made.
Also "no reason" is curious when it's the explicit pattern this exact industry has exhibited. Chinese models lag a few months, and come in swinging with an order of magnitude more efficiency. Does it apply? Who can say. It's certainly on the table. Pretending like it's any more ridiculous than AGI itself is irrational.
> Does this ever happen? Even in traditional manufacturing, the second “inventor” starts behind and has to improve their own process to surpass the first.
To use the industry itself once again: Japan was the first to deep learning. Performance issues prevented them from capitalizing on it. Now they're barely even a player on the field. Interestingly and adjacent to the industry, Japan was second to symbolic AI, and the FGCS was lightyears ahead of America's crufty Lisp ecosystem. For a more modern example, Google was the first to transformers. Much of this revolution is thanks to them. A shame that Gemini is third-rate at best.
It's not unique to this industry. That second inventor more often than not does a better job than the original one. It's a pretty common pattern. Otherwise, we wouldn't need patents.
My "contrarian position" is anything but. There is nothing new under the sun. You don't just need to be first, you need to maintain exclusivity.
xboxnolifes 5 hours ago [-]
Nobody argued exclusivity. But there have been clear advantages to having nukes first.
ux266478 4 hours ago [-]
> Nobody argued exclusivity.
And I quote:
> Imagine a world where only one country has AGI/ASI.
jr3592 9 hours ago [-]
As if time isn't money? Everything is temporary... Getting somewhere first has immense value.
ux266478 9 hours ago [-]
A reductive equation, economics isn't thermodynamics. Money is fictional and value is subjective and unstable. Within this context, being first to AGI means nothing if the second invention of it comes 2 months later and works an order of magnitude faster than what the first iteration had self-improved to at that point in time. First mover advantage isn't decisive, you have to actually be able to capitalize on it in a robust way.
fn-mote 7 hours ago [-]
> works an order of magnitude faster
This is the weakest part of the argument. Absolutely no reason to believe the second inventor will be 10x faster.
Does this ever happen? Even in traditional manufacturing, the second “inventor” starts behind and has to improve their own process to surpass the first.
799tppp 6 hours ago [-]
[dead]
xeromal 7 hours ago [-]
Russia invaded Ukraine specifically because they have nukes. Everyone is afraid to fight back.
vrganj 9 hours ago [-]
This presupposes AGI or ASI are a real, reachable thing. My read is they might be, but not as LLMs. Until there's a fundamental rearchitecture, I'm AGI-agnostic and given that view, it doesn't seem rational to bet the house on it.
We'll see.
Palpatineli 10 hours ago [-]
How about "being able to align ASI somewhat to your values"?
ben_w 10 hours ago [-]
True, but that isn't EU/US/China, it's OpenAI/Anthropic/Grok/ …/DeepMind (based in UK)/… DeepSeek
With a lot of Chinese nationals in American companies, and an American corporation owning DeepMind, and a lot of people very upset with all of them at the same time, this is very messy.
procgen 10 hours ago [-]
The only prize is control of the light cone.
tedggh 9 hours ago [-]
A race to the bottom.
7 hours ago [-]
bpodgursky 10 hours ago [-]
It's red queen. You stay alive by winning, you lose everything by losing.
ben_w 10 hours ago [-]
Not sure you get either outcome in either case.
Race dynamics increases p(doom) for everyone.
The non-doom scenarios include "utopia for all", and "power flows to investors, not citizens of whichever nation the winning model's corp. was registered in".
Independently, "oh look all the investors went bankrupt" can happen in both "doom" and "normal technology" timelines.
ChrisClark 10 hours ago [-]
Unless you manage to build a god, and keep it under control... okay, we're all going to lose
kolinko 6 hours ago [-]
And US lost the significant role in chip/pc manufacturing. But if a product becomes commodity or utility (which at least for now it seems is the direction), with little lockin, it's not a big deal.
I hope we (EU) don't waste money trying to train local models (which at least some people in Poland try to do), and tries to build our own chips - AI chips have different architecture than regular processor/GPU, and TSMC doesn't need to be winner in this new race.
And if not this, then smaller labs, harnesses and actual application.
pbkompasz 9 hours ago [-]
Not being a rat in the rat race is the real win.
missedthecue 7 hours ago [-]
The largest non-US non-China model is Russian which surprised me.
sgt 7 hours ago [-]
Are you saying Russia is winning over Europe in AI already? Dang, sanctions really don't work.
IlikeMadison 58 minutes ago [-]
GigaChat is russo-centric and is completely useless outside Russia itself, performs absolutely worse than Mistral once you start using English. Reminder that Russia's economy is smaller than Italy's alone. They are not winning over Europe in any single metric, let alone against the poorest country in Europe, which is not even part of NATO nor the EU, on the battlefield.
NicuCalcea 7 hours ago [-]
Largest by what metric? And which model is that?
missedthecue 3 hours ago [-]
Largest by total parameters. GigaChat 3.1 Ultra: 702B
thadt 10 hours ago [-]
Yeah? And here I've been a happy Transkribus customer for some time now. If there are better models or interfaces out there for analyzing historical handwriting, I'll definitely take a look.
deadbabe 7 hours ago [-]
Even Africa will surpass Europe with its massive data center upstarts breaking ground.
t3hTao 7 hours ago [-]
[dead]
onetwig 7 hours ago [-]
Huh? Mistral 7b was pioneering in its day and IMO they have been very on top of releasing niche useful models like moderation, OCR, etc.
I’m glad Mistral is working on useful solutions.
OpenAI/Anthropic is like a retarded little sibling chasing “AGI” and giving up on rich media and other modalities.
OpenAI/Anthropic is the worst of the mainstream AI.
It goes:
1. Gemini
2. Vidu
3. Le Chat (Mistral)
4. DeepAI
5. [insert MiniMax provider]
fn-mote 7 hours ago [-]
Totally fake post. Gemini in the top?? For what?
v3ss0n 3 hours ago [-]
Gemini is best at OCR and document extraction so far.
We had tested for insurance needs. None match Gemini even flash level yet.
zzleeper 2 hours ago [-]
Same here. Maybe Fable is better but in terms of cost effectiveness it wouldn't even make sense to test it
waldrews 7 hours ago [-]
The VLM's are so good at complex document understanding now. But you just can't trust them not to invisibly censor sensitive clinical/legal docs, even at the maximally permissive settings.
And the deep learning OCR-only models won't censor, but can and do hallucinate. I've yet to see a 'scan with different approaches and reconcile and say you're not sure if they don't agree' system just work for generic complex documents.
kolinko 6 hours ago [-]
The way my harness set it up is going through 2 or 3 providers, and cross-checking through them, also with plain text extracted if available.
I think we also had a layer that for any quote extracted tested it back if it exists within the original.
If you wanted 100% accuracy, I think it wouldn't be too difficult nowadays to guess the font&size&other text settings, and re render the crucial parts.
dangoodmanUT 5 hours ago [-]
> But you just can't trust them not to invisibly censor sensitive clinical/legal docs
What's an example of this?
merb 11 hours ago [-]
1000 Pages / 3.5€ this is expensive as hell.
If this is not fastly superior than something like tesseract it is not worth it.
beernet 10 hours ago [-]
Agreed. Does the GTM team there really sit together like "oh yeah, that sounds reasonable" while being totally beyond typical market prices?
Oras 8 hours ago [-]
Even comparing to AWS Textract or Azure Document Intelligence, this is very expensive (more than double)
piterrro 10 hours ago [-]
For anyone interested, I have an ocr pipeline running on rented GPUs, doing around 1000pages for 0.05-01 usd with around 0.8 seconds per page with full bounding boxes support for grounding.
If you’re interested you can find contact to me via this profile.
3.5 usd/1000 pages is just too expensive…
aliljet 10 hours ago [-]
Accuracy is truly what people die for in the OCR game. Price isn't the primary function here.. it's an equation of price, accuracy, speed, and in mayn cases regulation.
merb 9 hours ago [-]
Tbf even with tesseract you already get shit ton of accuracy and you can probably do these 1000 pages for way less than 3.5€. For 3.5€ you can spin up a cloud instance with 8vCPU+32gb on gcloud for 11 hours (or 11 instances for an hour) which can do way more than 1000 pages per hour on tesseract. It takes you around 6 second per page +-4 seconds start/stop depending on what you are doing on that instance size without too much optimization (you can probably even run multiple processes on a single node)
Google documentai costs 1.5$ per 1000 which is probably better in quality and speed.
kergonath 8 hours ago [-]
Tesseract is not a substitute for these models, which understand complex layouts and also extract bounding boxes for things like tables and pictures. They are also much better at making sense of cursive scripts.
I’ve been there, implementing a way to linearise text from a document with pages with 1, 2 or 3 columns, some of them in landscape is a nightmare. And that’s not even considering equations.
In the end it’s way easier to use a specialised model, trained by other people to do exactly what I need.
tjoff 8 hours ago [-]
Tesseract is super picky though often failing pixel-perfect screenshots...
So, maybe it can be tuned for your usecase but with that kind of investment €3.5 for 1000 pages is a bargain...
merb 7 hours ago [-]
It‘s also cheaper than the murican ones from Google, Amazon, …. And tesseract was an example. Heck you can go xberg and use paddleocr. Most often layout is less of a problem for ocr. Most often you need high accuracy, which tools like these are often worse in the 95 percentile.
v3ss0n 3 hours ago [-]
Are you serious? Tasserect is the worst compare to PaddleOCR , Surya/Marker
piterrro 9 hours ago [-]
Most use cases dont need that kind of accuracy, just doesnt justify the 3-4usd range. I build for that exact case (tender documents, we’re processing north of 100k pages per day), it doesnt need to recognize scanned written text from 1930s, its usually pdf/docs/scanned printed pages.
The accuracy is great, bounding boxes are must have for proper grounding for building answers by LLMs. Tesseract was too slow and not enough in some cases (for example tables or images which we also recognize and describe)
pogue 9 hours ago [-]
If you're getting inaccurate results from OCR what's the purpose of even doing it? Inaccuracy of text of any kind seems like a completely obvious failure of the entire purpose of scanning text into a computer.
fluoridation 9 hours ago [-]
It depends on what you need. For example a while ago I scanned and OCR'ed a bunch of receipts to get a timeline of my salary. I only cared about the gross and net figures, and nothing else mattered. Tesseract's output had a bunch of errors and misdetections, but the main figures always came out OK, and a local LLM was able to pick them out from the noise every time.
There's a big gulf between "it's as if a human being had transcribed it and reconstructed the original document" and "so completely broken it can't be used for anything".
piterrro 9 hours ago [-]
Accuracy can have different dimensions, depends on what you can tolerate and whether you can detect it to apply more powerful methods.
Imagine you have a cheap and 99% accurate ocr. The other 1% you can detect and apply more powerful (more accurate but slower and more expensive) ocr method.
What would you use? At scale these things add up.
Telemakhos 8 hours ago [-]
Can it produce accessible PDF files that will pass accessibility tests? Someone who can do that will make a killing laundering PDFs for academia: by April 26, every PDF, syllabus, and academic document needs to comply with WCAG 2.1 Level AA, which means structural tagging, alt text, and lots of other checklist items that AI could probably generate.
to21as 8 hours ago [-]
[flagged]
x3ro 10 hours ago [-]
You should put contact details in your profile :)
vrganj 9 hours ago [-]
Is it European-hosted and fully outside of both CLOUD Act and CCP reach?
Because I'm assuming that's why they get to charge more for the right type of customer.
piterrro 9 hours ago [-]
You can even run it on your desk if you want, a single gtx 4090 is enough. It can be fully air gapped.
ks2048 8 hours ago [-]
Does anyone know a site that lets you browse examples of input / output pairs?, particularly with layout analysis (bounding boxes of figures, tables, etc).
ianhawes 10 hours ago [-]
I won't comment on accuracy, but in internal benchmarks, Mistral OCR is significantly faster than comparable APIs.
maelito 8 hours ago [-]
Given the latest vibe release's new "follow default" model option, we should see a new coding / general Mistral model, mistral 4, soon.
oliveralbertini 7 hours ago [-]
I'm wondering if this model performs better on french (and other european languages) documents than others
Johnny_Bonk 10 hours ago [-]
How does this compare to Baidu Unlimited OCR. I've been very impressed with Baidu and it's essentially free to run on a decent computer, other than electricity costs.
spiderfarmer 10 hours ago [-]
Where do your documents go?
rescbr 10 hours ago [-]
They go to the decent computer hosting the model, which can be yours if you pay the electricity costs
ad_fontes 10 hours ago [-]
I've been experimenting with using NuExtract this week on locally OCRing bank statements that don't have a predefined document structure. It's way better than Tesseract or a generic vision-enabled model. It runs great on a single RTX 4090 at the modest throughput I need.
Their hosted, API-based service is something like a third of the cost of this model.
parhamn 9 hours ago [-]
Mistral is bumping the price of this thing every release. I think we're at 2x now?
maz1b 10 hours ago [-]
How does this compare to 4?
mainecoder 11 hours ago [-]
The chinese did it better, mistral is alive thanks to regulations.
Bombthecat 9 hours ago [-]
Yeah, I'm not sending personal bills etc to china. No thanks
gkbrk 8 hours ago [-]
It's still open-weight models that you can download and run locally. You don't need to send anything to China.
sajithdilshan 7 hours ago [-]
and you trust the french?
MrDrMcCoy 4 hours ago [-]
I trust them more than the Americans and Chinese. Genuinely rooting for them to close the gap.
maelito 9 hours ago [-]
No it's the other way round : the Chinese do better thanks to regulations : massive amounts of money from Big tech and public money.
mangecoeur 10 hours ago [-]
i.e. it's one AI company that's basically guaranteed to never fail since it has a market niche guaranteed by European companies and governments.
petcat 10 hours ago [-]
Which is also why their most recent model "Shieldstral" does nothing except monitor and moderate internet content.
After stuff like Chat Control I think they're obviously seeing a big demand for this kind of "internet safety" technology in Europe.
rtaylorgarlock 10 hours ago [-]
I've been a bit more careful about complaining about regulations broadly due to competitive advantage, e.g. ITAR
t3hTao 8 hours ago [-]
[dead]
hmokiguess 9 hours ago [-]
Whoever is paying all that for OCR is being scammed.
raverbashing 9 hours ago [-]
* scanned
tethys 7 hours ago [-]
Who the hell at Mistral thinks it is a good idea to register CMD + T as a shortcut for switching theme!?
Nothing special about this model for overly-detailed work like mine.
It's been a while since I last tested (and discontinued my subscription), but the "pro" models from OpenAI dominate. Not surprising, given the price difference, but it would be nice if an OCR-specific model could perform better. It's worth mentioning that even the highest-end models do a pretty poor job with intricate text like mine.
Mistral's one advantage is that Anthropic now flags OCR, because they don't allow anything that could be considered "reproduction", even of work for which you own the copyright. So my new workflow is Mistral OCR for the actual OCR, followed by a proofreading pass by Claude (which is allowed). Claude is obviously more expensive, but it caught entirely hallucinated sentences created by Mistral OCR 4.0, so I was glad for the backup check.
I feel Anthropic is destroying itself with all these restriction. They got away because their models were the best for coding, but that is not an advantage anymore as OpenAI and other open source are already better.
I mentioned it in a sibling reply, but here's Anthropic's support document about not using Claude to reproduce content verbatim that already exists, regardless of copyright.
https://privacy.claude.com/en/articles/10023638-why-am-i-rec...
What does this entail? What does Claude do to decide that the text it was provided was hallucinated? Are you telling Claude that the source was OCR'd by another LLM?
I can't speak for Mistral OCR 4.1, but the hallucinations in 4.0 were so egregious (just completely making up new sentences in the middle of a page) that I knew I can't trust Mistral OCR on its own.
I haven't seen any difference in my ocr workflows, what do you mean by this?
I’ve also had it refuse to OCR public-domain books that included content that it didn’t like, such as references to prostitution in 19th-century books about Japan.
I had one session where Claude refused to continue after it hit some kind of guiderail restriction. I couldn’t see what the trigger was, so I started a new session, gave Claude the link to the previous session, and asked it to diagnose the problem. This new Claude said it couldn’t view the exact guardrail issue, but it did suggest a workaround that turned out to be effective.
At first I thought it was something in the scanned content that was being flagged, but it was the attempt to transcribe that was itself being flagged. Anthropic mention it on their pages:
"Anthropic takes these steps because Claude’s purpose is to generate new content and ideas, not to reproduce content that already exists."
https://privacy.claude.com/en/articles/10023638-why-am-i-rec...
Side note - Claude itself is not aware of this policy, and is unable to see the API responses - the turn just ends. Which turned into a really bizarre failure state where Claude thought I was gaslighting it and kept insisting it could do the work and even had the entire text in memory. Every time it would go to show me and prove it, it would hit API Error 400. I was only able to convince Claude by showing screenshots of my Claude Code screen output so it could see that I was seeing API errors. I've never seen Claude get into that angry & snarky state before, and I hope it doesn't happen again.
That said, models have sometimes surprising weaknesses and a model could be terrible overall but magically work for one type of document.
I'd love to see some advancements in traditional OCR based on ideas and concepts we've learned from newer "OCR-like" models since traditional OCR is drastically cheaper.
I haven't been impressed with any of Mistral's models. They obviously realized that they couldn't compete at the frontier so they decided to go for smaller focused models but even those have not been that good.
We moved away from Cursor but I was looking for a model that would help with FIM (fill-in-middle) multiline autocompletion and people were recommending Mistral's Codestral. We gave it a shot and it was lackluster at best.. Even Google's Gemini did a significantly better job than Codestral.
Ultimately Opus-class models got good enough and I don't do much manual coding anymore.
I really wouldn't know, though. Anthropic models barf out copyright issues for my use case, so I'm unable even to benchmark them. It's a common problem when you're scanning public domain books. Mine are reference texts often cited.
As another user pointed out, it's surprisingly random (task-specific). Llama Scout outperformed Gemini Flash 2.5 on a benchmark I built at the time. I didn't include an OCR models.
Mistral might indeed be the best OCR-specific model for my task, now that you ask. Funny. It's so bad at my work that I didn't register it might be the best in its category. This is just based on vibes from my single scan.
"Winning the race" doesn't give you that in any meaningful capacity. It gives you, at the absolute most, a temporary window where that's the case. See: nuclear weapons.
Europe already has an innovation problem that's already causing structural economic instabilities which Germany has been struggling (and lately failing) to prop up.
As much as it pains me to say this, AI is already a tech revolution and it seems like Europe is just ignoring it. There's more innovation in 3 blocks in downtown San Francisco than the entire continent of Europe.
QED being first didn't grant exclusivity.
The premise wasn't that AGI isn't useful. There are actually layers to the metaphor where first-mover advantage of AGI is even less meaningful than it was for nuclear weapons, but I leave those as an exercise to the reader to discover.
This is your opinion. Lots of tech money appears to disagree.
You should at least try to explain your contrarian position, or cite your favorite source that makes an argument that you believe.
FWIW I explained it fully and provided the argument, along with explicit grounding examples. Can you tell me what exactly you don't understand?
To quote your other post:
> This is the weakest part of the argument. Absolutely no reason to believe the second inventor will be 10x faster.
Nor is there any reason to believe that it will work at any usable speed. An LLM capable of AGI running at 0.000001tok/s isn't a very good head start. Nor is it a good moat if it can't actually realize anything material fast enough. Lord knows America has labor problems.
I'm demonstrating the context is a lot more complicated than "be first". There are innumerable specific examples. They're not guaranteed to happen, that's not the point being made.
Also "no reason" is curious when it's the explicit pattern this exact industry has exhibited. Chinese models lag a few months, and come in swinging with an order of magnitude more efficiency. Does it apply? Who can say. It's certainly on the table. Pretending like it's any more ridiculous than AGI itself is irrational.
> Does this ever happen? Even in traditional manufacturing, the second “inventor” starts behind and has to improve their own process to surpass the first.
To use the industry itself once again: Japan was the first to deep learning. Performance issues prevented them from capitalizing on it. Now they're barely even a player on the field. Interestingly and adjacent to the industry, Japan was second to symbolic AI, and the FGCS was lightyears ahead of America's crufty Lisp ecosystem. For a more modern example, Google was the first to transformers. Much of this revolution is thanks to them. A shame that Gemini is third-rate at best.
It's not unique to this industry. That second inventor more often than not does a better job than the original one. It's a pretty common pattern. Otherwise, we wouldn't need patents.
My "contrarian position" is anything but. There is nothing new under the sun. You don't just need to be first, you need to maintain exclusivity.
And I quote:
> Imagine a world where only one country has AGI/ASI.
This is the weakest part of the argument. Absolutely no reason to believe the second inventor will be 10x faster.
Does this ever happen? Even in traditional manufacturing, the second “inventor” starts behind and has to improve their own process to surpass the first.
We'll see.
With a lot of Chinese nationals in American companies, and an American corporation owning DeepMind, and a lot of people very upset with all of them at the same time, this is very messy.
Race dynamics increases p(doom) for everyone.
The non-doom scenarios include "utopia for all", and "power flows to investors, not citizens of whichever nation the winning model's corp. was registered in".
Independently, "oh look all the investors went bankrupt" can happen in both "doom" and "normal technology" timelines.
I hope we (EU) don't waste money trying to train local models (which at least some people in Poland try to do), and tries to build our own chips - AI chips have different architecture than regular processor/GPU, and TSMC doesn't need to be winner in this new race.
And if not this, then smaller labs, harnesses and actual application.
I’m glad Mistral is working on useful solutions.
OpenAI/Anthropic is like a retarded little sibling chasing “AGI” and giving up on rich media and other modalities.
OpenAI/Anthropic is the worst of the mainstream AI.
It goes:
1. Gemini
2. Vidu
3. Le Chat (Mistral)
4. DeepAI
5. [insert MiniMax provider]
And the deep learning OCR-only models won't censor, but can and do hallucinate. I've yet to see a 'scan with different approaches and reconcile and say you're not sure if they don't agree' system just work for generic complex documents.
I think we also had a layer that for any quote extracted tested it back if it exists within the original.
If you wanted 100% accuracy, I think it wouldn't be too difficult nowadays to guess the font&size&other text settings, and re render the crucial parts.
What's an example of this?
If you’re interested you can find contact to me via this profile.
3.5 usd/1000 pages is just too expensive…
Google documentai costs 1.5$ per 1000 which is probably better in quality and speed.
I’ve been there, implementing a way to linearise text from a document with pages with 1, 2 or 3 columns, some of them in landscape is a nightmare. And that’s not even considering equations.
In the end it’s way easier to use a specialised model, trained by other people to do exactly what I need.
So, maybe it can be tuned for your usecase but with that kind of investment €3.5 for 1000 pages is a bargain...
There's a big gulf between "it's as if a human being had transcribed it and reconstructed the original document" and "so completely broken it can't be used for anything".
Imagine you have a cheap and 99% accurate ocr. The other 1% you can detect and apply more powerful (more accurate but slower and more expensive) ocr method. What would you use? At scale these things add up.
Because I'm assuming that's why they get to charge more for the right type of customer.
Their hosted, API-based service is something like a third of the cost of this model.
After stuff like Chat Control I think they're obviously seeing a big demand for this kind of "internet safety" technology in Europe.