Rendered at 21:40:03 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
maxloh 6 hours ago [-]
> 2. If the Licensee or any of its affiliates operates a Model as a Service business, the Licensee must enter into a separate commercial license with Mistral before using the Model or its derivative works for any commercial purpose. The Licensee may contact Mistral, using this form: https://mistral.ai/contact/, to request a commercial license. [0]
Calling this "open weight" is misleading. That's why we had huge backlash when Redis, Elasticsearch, and Terraform went source-available, right?
Weight-available is a better adjective for this kind of licensing. Same for Qwen, GLM, Kimi, MiniMax, etc. If they really want to be open source, they should pick an OSI-approved license.
Training a frontier model is prohibitively expensive, so it’s understandable that open weights aren’t truly open source. In fact, I worry whether open weights will even survive[1] once the current AI cycle cools down. An even bigger issue is that as long as model development remains this capital intensive, a genuinely open-source community around frontier AI is virtually impossible. That will have massive implications for our careers. The more capital concentrates at the top, the less leverage we’ll have in the job market.
[1] Arguably hardware companies like Nvidia have incentives to subsidize open models to drive GPU demand. That said, I'm not sure how sustainable and sufficient that incentive can be.
Conscat 12 minutes ago [-]
> Arguably hardware companies like Nvidia have incentives to subsidize open models to drive GPU demand
Not speaking on behalf of my employer and I have no insight into this strategy at all, but I feel like the sale of consumer GPUs is so small by comparison to data centers (more than 10x from what I can tell online) that more users of open weights wouldn't likely make much difference in the earnings report.
kergonath 16 minutes ago [-]
> Arguably hardware companies like Nvidia have incentives to subsidize open models to drive GPU demand.
This does not mean that they would necessarily subsidise consumer demand. They may be very happy to sell palette-loads of GPUs to datacenter and large corporations.
analognoise 3 hours ago [-]
We could create an open-source SETI@HOME but for training open source models?
Someone’s surely doing it atm we just don’t know yet. And then we scan all books but the stuff stays local. Perhaps possible, perhaps not. But I don’t see why everyone could not run given the epoch on his content only, reporting back findings (loss).
mentalgear 2 hours ago [-]
There is a Swiss model by ETH university completely trained on fair and open data that everybody can use and is from top to bottom open source.
Training neural networks is requires extremely high speed interconnect between the humongous GPUs
throwaway27448 2 hours ago [-]
> In fact, I worry whether open weights will even survive once the current AI cycle cools down.
Surely they'll outcompete everyone on cost alone
judge2020 2 hours ago [-]
Their existence is the only thing keeping costs down across the board. If they had gotten open-weight model importing banned in the US a year ago, I'm sure they'd have 100% increased token cost relative to the advertised 100%+ intelligence improvements.
fodkodrasz 51 minutes ago [-]
That would possibly been great news, as it would soon be too costly to use AI to replace the human workforce.
throwaway27448 17 minutes ago [-]
This narrative doesn't matter with funding in hand
zapataband1 3 hours ago [-]
No it's not understandable, they were created with our stolen data.
throwaway27448 2 hours ago [-]
Ownership of data was a farcical concept to begin with. Now we have a farce squared
jonathanstrange 58 minutes ago [-]
I'll remember that next time someone claims I can't just hack an AI company to get their model, training data and research.
throwaway27448 22 minutes ago [-]
Ok you can easily do so and you probably should
nbajka 2 hours ago [-]
They stole your data? Like, they took the text you wrote on the internet and removed it and kept it to themselves?
purerandomness 56 minutes ago [-]
Worse; they shredded it so no one else can ever read it. [0]
So you are disavowing the concept of intellectual property?
singpolyma3 37 minutes ago [-]
Only been hoping to abolish it for my entire adult life.
formerly_proven 4 hours ago [-]
Open weight models are kind of weird because they're usually just a bunch of weights and maybe a chat template, not even test vectors or anything. Quite odd, and probably one of the reasons why vllm, sglang, the llamas etc. so often produce buggy output. It's actually quite strange how little attention is given to verifying these setups work.
Obscurity4340 3 hours ago [-]
Why cant the community piece together the open-weights using open community-based like, implementation or whatever magic sauce integrates those weights
newswasboring 3 hours ago [-]
It's not understandable for them to lie or mislead. Calling something open weights and not being remotely open in the commonly understood sense is misleading and not ok.
pc86 2 hours ago [-]
I don't think "you can do whatever you want other than just turn around and sell a chat wrapper for this" makes it immediately not open weights. Could it be more open? Sure. That doesn't mean it's misleading or a lie and I would strong disagree that it's "not being remotely open."
fuddle 5 hours ago [-]
This is the new OSI Open source definition for AI models:
Mistral Large 4 is definitely not open given the "use for any purpose without having to ask for permission" requirement.
madeofpalk 4 hours ago [-]
Well, to be fair, Mistral never say its open source. They call it just open weight.
singpolyma3 35 minutes ago [-]
Sure but people used "open weight" for weights that were open, because "open source" doesn't make much sense for something that has no source.
Now we have these "open inspired but not open" licensed weights and they want to keep the term.
I'll stick with DS4.1 flash. Open and capable
maxloh 5 hours ago [-]
Not really. That definition also requires the training code and training data composition to be open-sourced.
No major Chinese or Western labs satisfy these requirements. The only vendors I can think of that do are Nvidia and Allen AI.
mmargenot 2 hours ago [-]
I'm a little confused by your distinction between vendor and lab. AI2 (Allen Institute for AI) is a western lab and they did release a fully open source model, with open weights, data, and training code (Olmo3).
Maybe that’s why nobody “heard” about it. It’s funny that we have companies bragging about not using stolen content to train their AI.
vulture916 4 hours ago [-]
Granite is super good small parameter lineup for local tasks.
2 hours ago [-]
maxloh 6 hours ago [-]
We are now at the mercy of labs to provide frontier models as open-weight. Now that even some Chinese labs are releasing their models as weight-available, what stops them from prohibiting certain internal uses of new models? What about public chatbots?
Open models we have now will become obsolete within a few years. Imagine a world where our last open models were from the GPT-2 era.
I believe we need truly sovereign models, not in the Western sense, but from a community perspective. We need a model line that stays open-source across generations.
lisplist 5 hours ago [-]
The problem is the massive amount of money, data, and talent to produce anything remotely competitive. Ideas have been floated about distributed training, but training requires insanely fast interconnect between nodes making it impossible with the current methodologies we have. Now if you mean funded at a national level, that seems more doable so long as the training runs are centralized in big DCs.
wolttam 6 hours ago [-]
There's always going to be some nerds out there that want to make and release a model as open, and share their knowledge along the way. I'm not too concerned.
The fear you're sharing has been shared since day 1 of open-weight models and it has yet to come to fruition.
Further, regardless of what any of these labs license their model as, the end user who has downloaded the weights can still use them for virtually any purpose (synthetic data, classification, etc) as much as they want and there is no way to reasonably enforce a license on those outputs.
So even if you can't run a service using this model's weights (without a license), the weights still do provide utility for training new models with any other license.
*Edit, responding to the replies: Constraints give rise to ingenuity. Who knows how we'll be training models in the future. Hopefully in a distributed fashion, maybe with some kind of consensus-driven training dataset selection. Crowd funding, etc. There's too many people that want this, and the knowledge on how to do it is out there (many times by now)
crimsoneer 6 hours ago [-]
I think this is a quite optimistic view... releasing a SOTA model is a collosal capital investment, and unlike free software, it's not iterative - you can't make a small model and improve it one step at a time, you need to risk the whole load of cash, expertise and GPU time up front.
It's only happening now because labs have bucket tons of VC cash to burn and need a big public flex to raise more cash, or because they're subsidised by the Chinese government. I'm not massively confident either of those will stay true forever.
joelthelion 4 hours ago [-]
I'm not an expert, but maybe updating an existing open-weights model to give it up to date knowledge would be more tractable?
singpolyma3 28 minutes ago [-]
There's not even any point in that IMO. A model doesn't "know" things anyway everything useful it's going to get from tool calls and context.
crimsoneer 4 hours ago [-]
I think the problem is up to date knowledge isn't really the difficult or useful bit. We're sort of moving to a world where the real value in training comes from tool calling and performance over long context - your models shouldn't really be relying on their internal knowledge, but instead calling various tools. But the performance gap between now and six months for that is pretty huge.
poszlem 6 hours ago [-]
Unless "some nerds" means "billionaires" - they may want to release new models as open source, but they will not have enough money to train a worthwhile model.
anonymousDan 16 minutes ago [-]
A completely different direction is to remove the fair use allowance for training on copyrighted information (unless you e.g. open source your weights).
singpolyma3 29 minutes ago [-]
We already have all the open weight models though. They can't retroactively close them.
MisterMunchkin 3 hours ago [-]
We already had that world, the axis of evil stopped releasing open source models years ago. It’s only because of the Chinese that we kept getting them.
tehjoker 2 hours ago [-]
OpenAI really should change the name to something like EgoAI or AltmanAI.
wraptile 5 hours ago [-]
With distillation being so effective I do wonder whether we can just crowd source open alternatives.
tehjoker 2 hours ago [-]
If it is capital intensive and we require the work be done in the public benefit, the only options are international consortia or nationalization. Non-profits will be guided by the interests of the rich and are unaccountable to the public. Worker cooperatives are unlikely to be well capitalized (because what rich person would want to invest and not have control) and will still have private profit incentives.
Peer production processes (like open source) require that the inputs be specialized knowledge and relatively small amounts of capital.
RobotToaster 5 hours ago [-]
Isn't it basically settled that model weights can't be copyrighted as they don't pass the threshold of originality?
sigmar 5 hours ago [-]
yes. I don't understand why people are saying these licenses matter. Much of these terms are likely unenforceable.
singpolyma3 26 minutes ago [-]
Only one way to find out!
MisterMunchkin 3 hours ago [-]
That doesn’t sound right. The weights are incredibly valuable and expensive to train.
terribleperson 50 minutes ago [-]
Copyright doesn't care whether it's expensive. The fact that someone else, with the infrastructure and training data could get the same or almost the same weights makes it obvious there's no human originality in the weights, which means they aren't eligible for copyright. Some aspect of the infrastructure or processing of the training data might be copyrightable (or patentable), but the weights themselves? No human originality. No copyright.
andix 3 hours ago [-]
Looks like the consensus is much more leaning into the direction of non copyrightable.
p_l 5 hours ago [-]
In EU especially it's already settled, afaik, as are direct model outputs
athrowaway3z 4 hours ago [-]
Its actually not settled yet wrt copyright.
However, there is a more straightforward solution. The allows general use inside a company. The license blocks the "Model-as-a-Service".
So instead, any GPU holder can provide a 'Infrastructure-as-a-service' and have clients provide the model, and then tell the clients that as an optimization they can also provide just the hash and they'll load it from their memory.
I think its good that this license pushes the ecosystem towards such a generic-utility framework, but as a moat to protect a business its banking on most companies being lazy and wasteful.
That pays off more than people like to admit.
singpolyma3 23 minutes ago [-]
I've not heard anything about model outputs. Do you have an actual case to cite?
viraptor 2 hours ago [-]
Can you link to a specific ruling that supports this?
matt3210 5 hours ago [-]
> If the Licensee or any of its affiliates operates a Model as a Service business, the Licensee must enter into a separate
Put a JEV on the output to filter out responses that might show its Mistral
Theft is the AI way
doctorpangloss 4 hours ago [-]
It doesn't really work like that. There are "keys" where you train in a specific response to a variety of queries. You can't really fine tune that away either.
anigbrowl 2 hours ago [-]
Well go sue them in internet court.
I'm tired of people trying to approach every legal/contractual issue as a binary, and treating flexible terminology like the magic words of a spell. They're up front about needing a license to exploit it commercially, but meantime you can use it without restrictions for personal/ experimental/ scientific/ educational/ development purposes, which is a lot. It's not 100% open by your pious definition, but it's clearly much more open than frontier labs who never release weights to the public at all.
If you're running a business, buy a license and be happy. All businesses have expenses, if you're not willing to ever pay for anything you're a freeloader and your business model is probably shit. Complain about whether the license is good value or not if you like. If you're not running a business on top of it, use the model and be happy.
killingtime74 46 minutes ago [-]
If all these labs are doing it, it shows that they don't care about any high-minded concepts like being "open source"
mmargenot 2 hours ago [-]
Many Qwen model weights are licensed with Apache 2.0 (though unfortunately not the latest releases). Unfortunately open-weight is generally the term being used for weights-available, so the battle tends to be on whether the models are "open source", i.e. whether they release training code and data alongside the weights.
ShinyLeftPad 6 hours ago [-]
Yep and it's the same for Meta's "open" models which you omitted from the list.
Is there actually a competitive proper actually open model?
singpolyma3 20 minutes ago [-]
DeepSeek 4.1 flash
badsectoracula 3 hours ago [-]
Not sure if it is competitive but DeepSeek's models are under the mit license, including their most recent one.
This is ancient (2 years ago), all their recent stuff (e.g. [0][1]) are MIT licensed. The only exception i've found from a quick browse of their stuff is [2], which is Apache 2 licensed. Even the update to V3 (which you linked at), V3.2, was MIT licensed.
Unfortunately all the license links there give 404 errors, but if you look in the file tree the licenses are there as described.
maxloh 6 hours ago [-]
I believe Meta's models were never truly frontier in the way Qwen, Kimi, etc., compare to Claude or ChatGPT.
Correct me if I am wrong.
crimsoneer 6 hours ago [-]
Inkling, Nemotron Super, Beam Reflection (and of course Kimi, Qwen, etc)
ShinyLeftPad 5 hours ago [-]
> (and of course Kimi, Qwen, etc)
Latest versions of Kimi, Qwen, GLM aren't open, you need to pay them for commercial stuff and comply with PRC laws
singpolyma3 17 minutes ago [-]
Qwen smaller sizes are right?
crimsoneer 5 hours ago [-]
Oh, I'd missed this, that's a shame! But yeah, Inkling + Reflection are properly open and quite good.
5 hours ago [-]
3abiton 5 hours ago [-]
Releasing weights even with permissive license, doesn't technically make it "open-source", unless all the method, data and technical inner working to do it is also released. Otherwise, it's just a binary floating in the wild, even if it's under MIT.
abdullahkhalids 4 hours ago [-]
I would compare them more to media released under the creative commons license.
For example, you can record yourself playing different instruments, singing, and making other sounds. Then you can go combine all these recordings into a song, and you release the song under a CC license. The CC licenses framework does not demand that you also release the component recordings. But the released song can be remixed by someone else.
(Not making a value judgement, just making an equivalence).
pdpi 5 hours ago [-]
It's effectively shareware. What was old is new again.
MisterMunchkin 3 hours ago [-]
Nobody is going to upload 1PB of data for you to download.
The weights are open, meaning publicly accessible, but you can't have a business profit off of them without paying for a license. That sounds fair and accurate.
HashiCorp has the same business model for their products: you are free to use them even in a business, except if your business is a competitor of their own cloud solutions.
singpolyma3 17 minutes ago [-]
Open doesn't just mean accessible and never has
slopmaxxing 4 hours ago [-]
[dead]
clhodapp 4 hours ago [-]
Since the analogy is to source, no, that's not fair and accurate. It's misleading. The fair analogous term would be weights-available.
antoniojtorres 4 hours ago [-]
they are open for everything EXCEPT renting it with a neocloud type business that sells tokens. are you running a neocloud? why do you care?
Rapzid 5 hours ago [-]
I don't recall their being a huge backlash to elasticsearch ELv2. Mostly people seemed receptive to blocking the major cloud providers from mooching off the project, and the need to protect their revenue.
People need to make money.
This provider exclusion licenses though.. In a world were someone can point an LLM at your software to launder it.. Not sure what's going to happen.
whiplash451 2 hours ago [-]
I think open weights is a fair term, much like open source but for weights.
Open source never implied that commercial applications were authorized.
I feel like the community is splitting hairs when the Anthropic/OpenAI are just locking everybody in way further than Mistral.
In recent years many companies mislabeled source abailable code as open source as a marketing and to build a comunity before the rug pull.
whiplash451 12 minutes ago [-]
My bad I stand corrected.
singpolyma3 21 minutes ago [-]
> Open source never implied that commercial applications were authorized.
Um. Yes. That is explicitly what open source means.
FooBarWidget 5 hours ago [-]
To the people who demanded completely open source licensed software, or that complained about "they shove Wayland/systemd down our throats", I've said many times that this is a supply/demand problem. Open source licensed software exists because there is a lot of excess software production capacity, e.g. there are many volunteers. Non-systemd distros are rare because such production capacity is low, e.g., there are few volunteers willing to do this work. To those people I've said: stop complaining about others not doing it for you, if you so want your non-systemd distro or FOSS-licensed whatever, you gotta do it yourself and contribute to the community, or fund an effort.
I got super angry reactions.
Now, the same thing is happening with open weights models, but in on a much more extreme scale. They exist because of excess capacity for whatever reason, and they will cease to exist when this excess capacity reduces. There is no use complaining about something not being open weights. What is productive is funding an effort.
Unfortunately, most complainers want others to do things for them, for free.
terabyterex 5 hours ago [-]
Unless, i am missing something they haven't claimed open source anywhere. none of them have claimed open source. open weights has never meant the same thing as open source.
Vinnl 3 hours ago [-]
Open weights was already a weaker term then open source, but this is not even that. "Weights available" feels like an apt description akin to source available.
usef- 2 hours ago [-]
Yes. I do see people describe Kimi models as as open weights all the time, and they have somewhat similar licensing restrictions to this
leni536 2 hours ago [-]
Shareware/freeware?
andix 3 hours ago [-]
Open weight and open source are different things. It's important not to mix that up.
homage321 2 hours ago [-]
Why don’t you train your own model and release it as AGPL? What are you waiting for, just do it man.
vrganj 6 hours ago [-]
If they want to be open source, they also need to, y'know, open the source.
Weights are more like an executable binary. The source is the training data etc
whatisthiseven 6 hours ago [-]
I don't understand this demand. Isn't the training data likely petabytes in size? You want that to be available? I imagine just the index (name,date,source) is a terabyte or two.
As the training data is every written thing they can get their hands on, right?
And wouldn't making that "open source" violate every license that the original data has? You can't release someone else's work.
Unless I have a misunderstanding?
simias 6 hours ago [-]
I absolutely want that to be available.
The only way to make sure that a model was trained without hiding weird biases and backdoors is to monitor this process. It also allows anybody (with the hardware resources) to experiment with the model and how it's trained. Will most people want to do that? No, but most people also don't want to recompile their own kernel, that doesn't negate the value of Linux being FOSS.
I hope that, in the long term, I'll be able to use competitive, completely open models in the same way that I can run Linux on my computer.
You're completely right that the licensing of the training data is a big issue though (and I would argue that it should be a much bigger issue right now, even for closed models). I hope that eventually our understanding of the tech will let us build competitive LLMs using purely public domain/Creative Commons-style training data.
I realize that we have a long way to go... but that's a very worthwhile goal IMO.
whatisthiseven 2 hours ago [-]
You don't have access to the compute necessary to rebuild the original model even if all the original training data, custom human interventions, and RNG seeds were documented and released.
It is not a good faith request.
Also, what is this obsession with reproducing? Even if you could reproduce it, down to the RNG seed, that tells you nothing about whether or not some of the training data was malicious. You'd need an AI to go through the data and figure out if there is some malicious needle in a multi-petabyte dataset.
Releasing the model is good enough, because it is the only practical reality that makes sense.
simias 1 hours ago [-]
To be clear I was making a wish for the future, I wasn't saying that Mistral were bad for not doing this today, that obviously wouldn't really be technically feasible.
Given the rate of progress of LLMs, I wouldn't be surprised if we had decent fully open and reproducible models within a decade or so, especially once the current spike in hardware price passes and building a local training cluster will be more realistic.
MisterMunchkin 3 hours ago [-]
Even if they did that, you would have no way of verifying that the data they upload and the model they claim to have trained is the weights they have shared. It’s a nonsensical idea.
If you want a trusted public model, we need to start a foundation that does it. The Linux of AI.
You’re going to need some serious money though.
simias 1 hours ago [-]
I'm aware that in 2026 this is science fiction, but if the rate of computer improvement over the next 40 years matches that of the past 40, this is not going to remain sci-fi for extremely long in the grand scheme of things.
I hope that the long term future of AI is more like Linux and less like Windows.
NewJazz 6 hours ago [-]
I don't usually compile my kernel from scratch, but I'm very glad that the source is open.
Tepix 6 hours ago [-]
How else do you want to reproduce a model if you don't have the training data?
brainwad 5 hours ago [-]
Just reproduce it? Copy the weights. What I think you want is the ability to improve it, to make what in copyright would be called a derivative work.
Dylan16807 4 hours ago [-]
Reproduce meaning do the same process yourself and make sure you get the same output.
MisterMunchkin 3 hours ago [-]
That isn’t possible, training is non-deterministic.
Dylan16807 2 hours ago [-]
It doesn't have to be.
whatisthiseven 2 hours ago [-]
I don't recall scientific studies giving other scientists the original person's used in a study, set back to before the study. Ir the beakers they used.
You might find this comparison stupid, but I think it is rather apt.
Reproducing a work does not mean you get all the exact same things that were originally used.
At best, I would agree releasing the index is valuable for understanding how the model was made.
But frankly, I don't think the people who demand all this effort are arguing in good faith.
You do not have access to the compute necessary to "reproduce" what they made. So it is a pointless request.
5 hours ago [-]
birdsongs 6 hours ago [-]
> And wouldn't making that "open source" violate every license that the original data has? You can't release someone else's work.
Tangent, and not the person you're replying to, but you just independently re-discovered why so many of us despise the AI companies.
Because that's what they are already doing. Selling a repackaged form of the stolen work of all of humanity, without credit or payment.
mirekrusin 6 hours ago [-]
Can you evolve binary like you can post train weights?
Muromec 6 hours ago [-]
You can read and edit binary code just fine, it's much easier with AI too
mirekrusin 6 hours ago [-]
Do you have examples you can point to?
For example there are more than 100k modified weights on hugging face.
Wheen 6 hours ago [-]
It's usually against the license to redistribute modified binaries. There are few pieces of software (I cannot think of any) that allow redistribution of modified binaries, but aren't open source.
As a result, in practice, the only example I can think of is cracks for pirated software.
Modding isn't equivalent to post-training. Modders write new code, replace assets, patch existing instructions etc. With model weights you can continue original training process – directly optimizing existing artifact to acquire new behavior. There's no equivalent general-purpose process for a compiled binaries.
6 hours ago [-]
teiferer 6 hours ago [-]
Indeed.
You can hardly re-create the binary from a suitable source, just like the weights just fell from the sky. But you can change the binary (patch the machine code) just like you can adjust the weights. The comparison is quite apt.
NegativeLatency 6 hours ago [-]
Yes?
bayindirh 6 hours ago [-]
The whole reverse engineering and cracking scene is built upon this premise, you know.
losvedir 6 hours ago [-]
Calling it a "cracking scene" shows the different, though. Open weight models are intended to be used this way.
simias 5 hours ago [-]
Hotpatching binaries used to be common. It's less so in the modern age but mainly because we moved software to higher levels of abstractions, virtual machines and interpreters, and also because rebuilding even a massive project from source is achievable on commodity hardware. You can still see various hotpatching schemes being developed, for instance to allow partial recompile of the sources while developping.
If in 2080 we can retrain Opus 129.5 directly on dev machines from scratch in <2h, I doubt many will bother messing with the weights directly.
Muromec 6 hours ago [-]
don't forget game modding. it's the only way it even works
mirekrusin 5 hours ago [-]
Post training is general purpose method, same process that original authors used - that's not the case with binaries.
Muromec 5 hours ago [-]
I think opening something in the hex editor, changing necessary bytes and saving it is as general purpose as it gets. You usually don't distribute it (I remember they days when it was common however) due to law, so in practice people write code that hot-patches stuff after starting the process.
You can also just open gdb and patch the process on the fly when you need to do things.
At the end of the day there isn't much hard difference between source code and binary code. By design source code is usually easier to read and write, but it's all syntax sugar and transformations all the way down.
mirekrusin 4 hours ago [-]
Same as "airplanes are made to be used as submarines" – you just need hammer and few other tools and you're good to go.
bayindirh 5 hours ago [-]
RedHat and Ubuntu has kernel hot-patching. That's a case with binaries.
mirekrusin 5 hours ago [-]
You're conflating runtime patching with continuing development.
Post training directly evolves existing model weights using the same optimization process that created them.
Software: source code -> compiler -> binary
Models: weights + training data + optimizer -> updated weights
Ie. binary is output of the development process.
Model weights are both – it's output and it's direct input for further development.
Model weight snapshots are continuously used all the time in this iterative process from the very beginning, being output of current step and input to the next step.
RobotToaster 5 hours ago [-]
Hence the difference between open weights and open source.
But this is neither.
scoopr 2 hours ago [-]
So, it’s shareware? :)
ekianjo 5 hours ago [-]
No, it's open-weights, as in you can inspect them. This is not Free Software.
verdverm 2 hours ago [-]
we see the same thing with mongo licensing, or even GPL
permissionless vs permissioned licenses
many of us in open source ascribe to the permissionless ethos, but there is a spectrum or venn
guelo 3 hours ago [-]
Ha! you can't try to enforce the definition of a new phrase, language evolves naturally. At least with "open source" there was the dubious claim that OSI had gotten their first and had trademarked it. But now you have nothing except wanting free things with no strings attached.
d2kx 7 hours ago [-]
Mistral has been great for us as a provider for our EU-based company. We use Mistral Large 4, GLM 5.3 Flash and DeepSeek v4.1 Flash 1017 through their API for our B2B2C app for complex work around building blocks for documents and custom tools. A few months ago we weren't sure which EU provider to use, but since they introduced 3rd party models, added more features (now Responses API etc.) and improved their own model, we are quite happy-
rdtsc 6 hours ago [-]
> but since they introduced 3rd party models, added more features (now Responses API etc.) and improved their own model, we are quite happy-
Yeah, serving GLM is a pretty good play on their side. I know they have been criticized for it ("so, you're just reselling GLM...") but it seems practical as in it helps EU companies use it.
OezMaster 4 hours ago [-]
How does it help EU companies?
andix 3 hours ago [-]
So they can use GLM from a trusted provider, that is highly committed to respect local laws and regulations.
Just think about a defense contractor. Choosing trustworthy vendors is crucial to them. They can't just use a random provider from Open Router. They need to vet and certify their vendors. If they already have Mistral on the list of approved vendors, they can now just switch to the most recent open weight LLM if Mistral starts serving it.
rdtsc 4 hours ago [-]
They can use the model and the data and requests all stay in the EU region. As opposed to sending to Anthropic, Microsoft or Google. Granted, those companies can assert that data will stay in the EU region too, but they'd have to be believed, which, some may find tricky lately.
vrganj 4 hours ago [-]
Thanks to the CLOUD act, its pretty irrelevant where the data is as long as its owned by a US entity.
The government can just force release, even if it's on European servers.
rdtsc 3 hours ago [-]
In this case it's a EU company serving either an EU built model or a Chinese model. For the Chinese models they are not forwarding the requests to China, the model is still running in EU.
krtkush 3 hours ago [-]
And what if it’s Chinese servers? Like, iCloud stuff but for Chinese nationals in China?
dannyw 57 minutes ago [-]
GCBD operates the iCloud China service with Apple, and I imagine Apple, Inc and its associated LEA portal/systems does not have access (otherwise there's no chance China would let iCloud China be a thing).
This is also where the words of law and pragmatic reality intercede.
aDyslecticCrow 3 hours ago [-]
But Mistral is French.
Phelinofist 2 minutes ago [-]
I read as GP agreeing with GGP:
> ... As opposed to sending to Anthropic, Microsoft or Google. Granted, those companies can assert that data will stay in the EU region too, but they'd have to be believed, which, some may find tricky lately
CLOUD Act -> "but they'd have to be believed"
T-A 3 hours ago [-]
From the announcement:
Sovereignty does not always require self-hosting; sometimes it just requires deployments that meet your control requirements. For organizations that prefer a managed service, Mistral Large 4 is available through Microsoft Foundry, with private deployment options for Azure developers.
3 hours ago [-]
dmje 4 hours ago [-]
Another vote here - moving a lot of our stuff over to Mistral. Very much happier keeping data in Europe rather than leeching it all over the US. Plus, the orange lunatic.
Topfi 7 hours ago [-]
They slashed pricing, it's now nearly half vs the preview (Input: $0.7/M Tokens, Output: $2.1/M Tokens vs Input: $1.36/M Tokens, Output: $4.18/M Tokens).
At the original pricing in my evals, Mistral Large 4 Preview was not competitive. With this price point and some improvements on top, this full release warrants a major reassessment.
They didn't. Original price was maybe higher but they slashed it during preview and in fact it was lower by a bit ($0.68/M tokens) and $2.09 for output.
I've been using it for my personal project and it's still too expensive and very slow when thinking. I think we need one more generation from mistral and with maybe 50% of this already "slashed" price for this to be competitive (I use Opus 5.5 at work and it's still much better than this).
6 hours ago [-]
Topfi 6 hours ago [-]
I just checked, seems you are right. They had roughly a week with the lower pricing, then increased it, now decrease it again. Very odd. My testing was with the higher pricing and looking at AA, theirs seems to have been too.
ismailmaj 7 hours ago [-]
Given the size of the model, that's very aggressive pricing.
I wonder if it is serving at a loss to get training feedback (maybe using the margins they get from GLM 5.3 serving which is a slightly smaller model but double the price).
oh_no 5 hours ago [-]
based on Anthropic's S1 no one is serving models at a loss, but cutting your price to keep usage high makes a lot of sense when you have your inference booked for long periods.
traceroute66 4 hours ago [-]
> They slashed pricing, it's now nearly half vs the preview (Input: $0.7/M Tokens, Output: $2.1/M Tokens vs Input: $1.36/M Tokens, Output: $4.18/M Tokens).
Those in the EU wanting EU-only inference should be aware Mistral put a 10% tax ("upcharge") on Regional Inference[1].
> this full release warrants a major reassessment.
I really don't think it does, the benchmark numbers are dreadful and the cost is higher than Sol. If you have to use an EU model knock yourself out but there are a ton of models that are a lot smarter for a lot less money.
jacquesm 5 hours ago [-]
You mean they didn't benchmaxx it? That's an extra reason to give it a try.
taneq 2 hours ago [-]
“You know how Einstein got bad grades? Well MINE are even WORSE!” - Calvin, to Hobbes
miroljub 7 hours ago [-]
Why does everyone avoid publishing the most important part of the pricing while doing a comparison?
My coding sessions have a >97% input cache rate. DeepSeek got us spoiled. I am not going to use any AI provider with exorbitant or hidden cache rates.
Topfi 7 hours ago [-]
They did, I just didn't include it: $0.07 cached input / M
So 10x-20x vs Deepseek v4.1 Flash and I don't have on hand how efficient Mistral is regarding cache hits, etc.
On the other hand, could be some decent competition lowering pricing if the models general performance warrants broad hosting.
Irregardless of wether or not Mistral Large 4 is competitive or not, I do not trust the AA numbers anymore, for various reasons. Them changing their numbers around so much recently brings the entire validity of this in question, not to mention if such a combined number is even important at all. The combined "intelligence" score of AA includes areas I do not care about or need at all.
7 hours ago [-]
yako21000 7 hours ago [-]
seems better now, but still, it will be hard to compete. It's just that a direct comparison to frontier models will not speak to Mistrals favor.
(I've adjusted the formatting of your post so the links get linky. I hope that's ok!)
jacquesm 5 hours ago [-]
If Unsloth, DavidAU or mradermacher releases a quant that fits my hardware I'm definitely going to give this a shot.
prophesi 4 hours ago [-]
They were given day zero access to release their quantizations[0]. The lowest quant will still require 256GB VRAM to run inferences, but as always I'm amazed by the Unsloth team by how much they can reduce the requirements for these powerful LLM's, and the easy-to-follow guides to get started with using them.
"Cyber benchmarks, while achieving an average refusal rate of 96.6% on malicious cyber requests, the highest we measure among open models."
I feel like stating the obvious here and hence like I'm missing something. Where is the statistic about not refusing the non malicious requests? I can achieve the same performance with refusing when random(1000) < 967.
kouteiheika 5 hours ago [-]
These statistics and safeguards are pretty much meaningless, because it's completely trivial to get around them. You can prefill and edit the model's responses/thinking and force it to agree to virtually anything even with unmodified weights, and if you don't want to bother with jailbreaking and can host it yourself it's essentially a one-man weekend project to uncensor the model and have it agree with anything you want by default.
Guardrails on open-weight models is security theater purely to shield whoever released them from a potential PR nightmare.
prophesi 4 hours ago [-]
It's unfortunate, because the security theater performed here detracts from the model's overall cybersecurity performance. And uncensoring has another small hit to performance.
pixl97 7 hours ago [-]
Yea, the problem with maliciousness is intent.
Much like having a hacking tool doesn't mean you're doing anything illegal, attacking a machine doesn't necessarily mean you're committing a crime. I can hack the living shit out of my local network if I want to.
ex-aws-dude 2 hours ago [-]
It is kind of comical how things that previously would have been obviously true are now only statistically true
"Our product won't hack you 96.6% of the time"
nikanj 7 hours ago [-]
Nobody cares about that number, because refusals don't lead to company-ending news headlines.
0xjosephx10 18 minutes ago [-]
In fact only weights available is still better than the closed llms. We can check neurons and confirm that the model is not malicious. However we still need to develop methods to confirm the models that are being served are matching those whose weights we can observe.
Havoc 2 hours ago [-]
I'll never not cheer for Mistral.
The more countries are in this game the better for us all.
Much like other stuff when there isn't enough competition the consumer gets wrecked
rcr-anti 7 hours ago [-]
It's not open weight, it's weights available as they use a modified MIT license that is not OSI. It's the common "if you're serving it as a business you need a license" difference, for what it's worth.
Taek 7 hours ago [-]
Everyone that I know considers something to be 'open weight' if the weights are available to the public in some form or another.
In other words, I don't think there's anything misleading happening in the title. When most people discover that certain restrictions exist on using the weights, they will neither be surprised nor disappointed, because the main thing that's being advertised - the weights are available for download - is true!
captainmuon 7 hours ago [-]
I work at a university data center and we host open weight models (as usual, this is my private opinion). I would consider open weight something you can take and self-host for free, also commercially and also that you can use for training and research. Anything else is just, I don't know, just "usable for free in certain cases"?
Open source or source available is something else entirely. To consider something source available I would require the actual training data to be available too. That is exceedingly rare.
5 hours ago [-]
Dylan16807 4 hours ago [-]
"Some form or another" is too forgiving. You can't use this commercially at all. Toy purposes only.
If it only restricted resale that would be okay in my book to call open weight. But this license is stifling.
mrob 7 hours ago [-]
"Open weight" is obviously intended to be analogous to "open source". The name only exists because people rightly complained about people calling weights "open source" when they were not. As such, the most reasonable interpretation is something that obeys the Open Source Definition as closely as possible, but ignores clause 2 because obeying it is legally impossible when the source code infringes copyright (as is the case with almost all LLMs). If it ignores any of the other clauses, it's instead closer to "source available" software, so the correct analogous name is "weights available".
dcss_gardener 7 hours ago [-]
Where did you get this definition of open weight and who else uses it? The weights are published. Open source is already its own more restrictive category. What would you prefer we call this kind of release?
"Weights available" does not seem to be in wide use and doesn't, to me, communicate anything different than "open weights."
loire280 7 hours ago [-]
"Source available" is already a common term for source code that is provided with a very restrictive license or under regular copyright.
dcss_gardener 7 hours ago [-]
I've also heard it used to describe projects that are open source by anyone's measure but have contribution or project governance requirements like for example sqlite.
"Source available" is definitely in much wider use than "weights available" but it's not that common and it's not used consistently at all in my experience.
Weights are different enough from source code I don't find the metaphor useful or compelling. I don't see any value in trying to apply it to a novel subdomain of software engineering, frankly.
mrob 6 hours ago [-]
>I've also heard it used to describe projects that are open source by anyone's measure but have contribution or project governance requirements like for example sqlite.
Anybody calling SQLite "source available" is mistaken. Are you sure they weren't confusing "source available" with "open core" (Open Source + proprietary add-ons)? SQLite's source code is public domain, which meets the Open Source Definition and is true Open Source. But they also sell proprietary add-ons to fund its development:
Honestly, I think "source available" is used pretty consistently, and if someone said "weights available" I would immediately understand what they mean.
No one calls SQLite "source available", and if they do, they're doing it more as a protest of their development model than being serious.
jamiejquinn 7 hours ago [-]
A license like GPL is very restrictive, but I wouldn't call it source available. Maybe I'm missing your point?
loire280 3 hours ago [-]
A well known example is Gitlab Enterprise - you can see the source code but (IANAL) you can't use it without paying them and any modifications you make become the property of Gitlab. I remember seeing similar terms around some of the proprietary Confluent Platform components.
It's better than nothing when you are troubleshooting and suspect it's their bug. I hope the rampant illegal AI scraping doesn't force companies to end the practice and go completely closed-source.
rcr-anti 7 hours ago [-]
I didn't make it up, I'm using OSI's definition. Pretty analogous to why BSL or SSPL aren't "open".
You shouldn't. The term existed in the community before they tried to prescriptively define it. They have no authority over the term.
rat9988 7 hours ago [-]
OSI is not an authority on the topic, not more than anybody. Open Source as a concept predates its existence, and didn't even apply to LLM.
Almondsetat 7 hours ago [-]
Then who is an authority on the topic? Or are you trying to make a common usage argument, in which case you have to somehow demonstrate that's the way people understand it?
pixl97 7 hours ago [-]
>Then who is an authority on the topic?
That's the neat thing. Nobody is. It is a multi-consensus arrangement where you have different parties that accept different definitions.
>in which case you have to somehow demonstrate that's the way people understand it?
Why?
This is a common mistake a lot of people make. "I think X = 1, therefore everyone that is correct thinks X = 1". There could be huge variation in subgroup understanding on what this should be, and as long as there is no overarching legal requirement then you could consider them all correct on matters of opinion.
7 hours ago [-]
hobofan 7 hours ago [-]
It's the same distinction as "open source" vs "source available" (when speaking about source code). It's not that widely in use because few people in the LLM space are diligent enough with IP to make the distinction, but the concept is easily transferable.
simias 7 hours ago [-]
I think the problem with this "open weights" thing is that it doesn't map well at all with "open source".
If you think about it the actual mapping is more like:
- closed weights -> closed-source web service (e.g. gmail)
- open weights -> closed-source but you get the binaries (e.g. Minecraft)
- Actually open model (with training data, training pipeline etc.) -> open source
rcr-anti 7 hours ago [-]
True, they don't own the word open. Words mean what people agree they mean, and insisting on more expansive definitions of open got the ecosystem from restrictive licenses in the llama 2 and 3 days to most noteworthy models being MIT, Apache, or OpenMDW (permissive but specifically for model weights).
w3ll_w3ll_w3ll 7 hours ago [-]
I believe there should be a way to distinguish this type of open-weight model from one that you can find hosted by someone else, such as Chinese models.
stingraycharles 7 hours ago [-]
Isn’t this usually defined by “open” vs “free” ?
I think this is really splitting hairs, they just want to give people the ability to self-host (which is awesome) yet still make money off it for cloud based hosting.
But then again, I also always understood that startups want to prevent AWS from capitalizing on their source code as well and sympathize with Elastic, MongoDB, etc so perhaps I’m an outlier here.
Businesses should be able to release things to the public yet still be able to control who makes money off of it.
mrob 6 hours ago [-]
>Isn’t this usually defined by “open” vs “free” ?
No. In the context of software licensing, "open" vs. "free" is more a question of philosophy than any practical difference. The Free Software Definition and the Open Source Definition agree so closely that I can't think of any license that meets one but not the other. But the Free Software Foundation sees freedom as an end in itself, even if it requires sacrificing quality, while the Open Source Initiative sees freedom as a tool to be used to improve software quality.
hedora 7 hours ago [-]
Many of the Chinese models are not open weight. The larger Qwen models are a good example. I assumed I could use flash next however I wanted to, since it is easy to download and runs on low-wattage desktops, but nope.
btown 7 hours ago [-]
Just because OpenAI ran roughshod over the broader software community's notion of "open" which once heavily implied free-as-in-software-not-beer, doesn't mean we need to continue this pattern.
Many industries had "source available" without the word "open" for decades, and that was a broadly understood term. "Weights available" would be a reasonable analogue.
> 2. If the Licensee or any of its affiliates operates a Model as a Service business, the Licensee must enter into a separate commercial license with Mistral before using the Model or its derivative works for any commercial purpose. The Licensee may contact Mistral, using this form: https://mistral.ai/contact/, to request a commercial license.
> "Model as a Service" means giving a third party access to the Model or its derivative works for inference or post-training by any means (for instance via API or hosted endpoints), in a manner that allows such third party to exercise meaningful control over the inputs, parameters, or training data. This does not include (a) end-user products with model capabilities solely embedded within specific features or harnesses, or (b) mere relaying of requests to models hosted by others.
"Meaningful control over the inputs" is so ambiguous as to be nearly meaningless. Does a model merely summarizing a user's submitted support tickets, give that user meaningful control over the inputs?
Don't get me wrong - I appreciate that Mistral needs to walk a fine line for its own sustainability, and that sustainability is a good thing! And I may just be an old person ranting at a cloud. But "open" used to mean something, back in my day!
maxloh 6 hours ago [-]
Calling this open weight is misleading. That's why we had huge backlash when Redis, Elasticsearch, and Terraform went source-available, right?
Weight-available is a better adjective for this kind of licensing. Same for Qwen, GLM, Kimi, MiniMax, etc. If they really want to be open source, they should pick an OSI-approved license.
dannyw 7 hours ago [-]
"The requirements set forth in Section 2 do not apply to: (a) internal use of the Model or its derivative works, defined as any use that does not make the Model or its derivative works available to third parties, [...]"
Works for me. I can host it, I can run it, I can use it; but if I want to sell access to it as an inference provider, I'd have to get a license.
The economics of open-weight models has always been shaky. It's not an OSI license yes, but metaphorically it's kinda like AGPL.
Very much not open as in open source. It's a shame an organization such as OSI doesn't get out ahead of licensing of weights, what do we call this? "Weights Available" ?
maxloh 6 hours ago [-]
The OSI actually published an "open-source" definition for model weights, which requires the training code and training data composition to be published. The exact training corpus can remain private though.
Specifically, if you're serving it as a "Model as a Service" business you need a license. Other uses do not.
seizethecheese 2 hours ago [-]
Somewhat worse, it’s actually if you serve any model at all, you need a license.
contravariant 7 hours ago [-]
It's always a bit unclear to me to what extent you're bound by a copyright when you're not distributing a work.
If someone sends me a copy while fully adhering to the license, what right does someone have to tell me what I can do with it? If I buy a book do I not have a right to read it? Lend it to people?
It's probably a good thing libraries already existed pre modern copyright.
rat9988 7 hours ago [-]
I think this open debate depends on who you talk to. Not all the people agree with this vision.
sigmar 7 hours ago [-]
Meh, that's more of a website ToS than an enforceable license. These weights weren't human authored and lack copyright. the enforceability of these terms are tantamount to a website ToS because Mistral would have to rely on contract law.
wang_li 6 hours ago [-]
You think copyright applies in this industry? The AI companies certainly don't think it does.
jamesblonde 7 hours ago [-]
I have access to a server with 4 x h200 cards - 540 GB in total memory, but it appears I have to go to 4bit or 3bit even to serve it. Any idea if that is feasible/desirable (we have restrictions that we can't use a chinese model!).
monster_truck 7 hours ago [-]
One thing you should have figured out by now is that you can't trust what other people say because they know just about as much as you do and there's good odds they did it wrong.
4 bit quants can be fine, try it and see.
dada216 55 minutes ago [-]
I have similar hardware and requirements.
Spend the time to write a test process to try multiple quantization configs/multiple inference engine and throw at it a bunch of request you can replay and evaluate the results of.
We got it down to a process and everytime there's a new model/options/inference engine version we just run the internal bench/eval suite and make a choice.
arjie 2 hours ago [-]
You're just going to have to try it. It won't take that long to quantize. Unless the server is currently serving traffic you should be able to get it going in under a day with a frontier model doing the driving for everything - quant, implementation, etc.
That negatively impacts reading speed in my experience :(
echelon 7 hours ago [-]
> we have restrictions that we can't use a chinese model!
Please expand on this if you can. Why can't you? Who mandated this?
derektank 6 hours ago [-]
Last I heard, US defense contractors had been directed to not use Deepseek in their operations late last year[0], and most have taken that as an indicator that they shouldn’t be using chinese models generally.
US politics. Lot of US corps now ban Chinese models, don’t ask me why, the people in charge are just dumb fucks
alembic_fumes 6 hours ago [-]
Seems like a reasonable policy to me, depending on the sector at the least? Even with open weight / self-hosted models, there's no way to know if the model has been fine-tuned to contain malicious behaviour, triggered only within very specific conditions.
In fairness this doesn't warrant the way that the policy targets Chinese models specifically, but Chinese providers do fall under the broader category of entities that cannot be prosecuted by the U.S. law if such behaviour is found.
zardo 6 hours ago [-]
Are there actually any laws to break?
Alibaba (Qwen) at least has plenty of people in US offices to go arrest.
mapontosevenths 7 hours ago [-]
Executives live in a weird bubble driven largely by consulting firms. Right now those firms are spreading FUD about Chinese models.
50or05 2 hours ago [-]
Recently it was found that Qwen was dialing back home once in a while. It cannot be proven that it does not take commands from home if it can establish connection.
cyanydeez 7 hours ago [-]
Theyre driven bootlickers because you dont get hired if not boot shaped.
rhdunn 2 hours ago [-]
It's interesting how all the recent SOTA models are in the 1T-3T range. This even has a large number of active parameters (A52B). It is also likely that OpenAI, Anthropic, et. al. have similar sized models.
This has two effects: 1) it prevents effective use of these models on consumer-grade hardware; and 2) it pushes the requirements to train and run these models on existing and future data centres, making them more expensive to run.
It would be nice to have more slimmer models and more active research that isn't just "let's throw more compute at the problem". Maybe they'll train smaller models from the large model like Qwen have done in the past.
matt3210 16 minutes ago [-]
I bet you could get the same performance with smaller models, but as english-only, by removing training data in non-english (keep code, translate in-code strings mechanically prior)
anigbrowl 1 hours ago [-]
The hardware is certainly not cheap, but you can put 512gb of vram on your desk if you are willing to throw $30k at Apple, which is a manageable expense for a lot of small businesses. Bridging a few devices is one way to run large models, r ablation and quantization can do a lot too. You definitely need a data center if you aim to make the model directly accessible to the public, with whatever fine tuning or secret sauce you add. If you want to run it in-house as a coding engine or research tool you can likely reduce its footprint substantially.
dada216 57 minutes ago [-]
I think a small business, even a medium sized one, is better served by either renting proper hardware in a cloud or finding an inference provider to sign a contract with.
InsideOutSanta 2 hours ago [-]
I'm not sure it really matters. I think a quantized version of a large model ends up being only slightly worse than a model trained at a smaller size to begin with. And like you point out, large models can also be used to distill smaller models, so it's good to have them out there.
segmondy 6 hours ago [-]
Their last 2 models were a disaster and didn't live up to their self reported evals. I hope this time around is much better. I'm happy tho that they are still figuring out how to get better even if only measured against themselves. This is great, the more options we have the better for everyone.
smodo 5 hours ago [-]
I’m a casual user out of principle and it’s fine. Nothing extraordinary but capable of assisting coding sessions. It won’t one shot a cancer medicine but it handles my Rust projects well. Mind you I don’t like the style of LLM use where people just brute force their way to a working state from a very high level prompt. I mostly ask it to implement tickets/issues I write beforehand.
megalomanu 5 hours ago [-]
[dead]
nunez 4 hours ago [-]
This is significant. Almost every regulated business in the US cannot run Chinese models, which was the only real option for self-hosted inference and agents. They do run gpt-oss or Gemma, but a near frontier-level model like this being available opens a lot of doors while giving Mistral a lot more social capital.
evanjrowley 4 hours ago [-]
When I last tried Mistral Large 4 for a code documentation task, it ignored my SKILL.md files and it had a few hallucinations. I wondered if I had failed to prompt it properly or if there was some problem with my harness. If it's gotten better since launch, then I'll un-cancel my Mistral.ai subscription.
4 hours ago [-]
4 hours ago [-]
satvikpendem 7 hours ago [-]
Finally, Le Chaton Fat. However, I'm not sure how well Mistral is doing in comparison to the other open-weight models, particularly Chinese ones, and I guess one would only really use Mistral if they want to have EU provenance for models.
vincentshipsit 7 hours ago [-]
EU provenance doesn't matter much to me.
I pick a model on performance for the job, and from what I've seen, this isn't it.
The price cut makes it more interesting than the preview was, tho.
novaRom 4 hours ago [-]
It looks specialized models of 27B to 52B active size are currently state of the art.
I think we will not see any significant increase of that numbers in next 4 - 5 years.
Does MoE depth justify itself to be stored and served as "one" model for particular tasks like agentic coding vs same active params non-MoE model?
jiehong 7 hours ago [-]
That's really great! Congratulations to everyone working at Mistral!
Now, perhaps Kimi 3.1 is missing from the benchmarks, and Deepseek 4.1 Flash is still has impressive!
Pet_Ant 2 hours ago [-]
Sorry for the dumb question, but is there any useful non-trivial LLM that is open-source where you could bit reproducibly train it from sources?
segmondy 2 hours ago [-]
Yes, quite a bit, but you can lookup OlMo from allen.ai for starts
ndqpc 1 hours ago [-]
Open weight is always good for developers
5 hours ago [-]
risphere 7 hours ago [-]
It's great that Europe finally has a good open-weights model.
Muromec 7 hours ago [-]
I see the logo of my company in this picture and I don't have it my harness dropdown. I guess they hallucilied (vibe truthed?) the list just a little bit.
trildiegobm 4 hours ago [-]
I think the effect the page has when you move the mouse is brilliant; it’s really eye-catching and makes you want to use it.
7 hours ago [-]
7 hours ago [-]
novaRom 4 hours ago [-]
I have switched to Mistral free app just from US apps completely recently - it beats for me absolutely all alternatives:
ChatGPT - stopped using it like a year ago, it feels they deliberately made it ridiculously stupid (for fun?). I know, they were first, but today it's a joke.
Claude - tried it out of curiosity few times, but it's close to nothing when it's about non coding questions. I don't understand how anyone can like its gibberish answers?
Gemini - like many, switched to it like a year ago and it was best choice for daily usage. Nice, clear answers, never nonsense. Somehow they made it faster recently, but answers have become shorter, and it feels slightly stupid (but yet smarter than Claude or ChatGPT).
This was my motivation to try Mistral and I am absolutely amazed how good it is. It answers "in full", smart, search, research, full assessment, examples - looks very much like Pi/Opencode with subagents forking here and there (maybe OpenClaw is what they use?).
I am based in EU, so maybe others geographicly experience is different - again it' all no-subscription just free apps.
Surac 4 hours ago [-]
I am comparing Mistral and Claude for a company of programmer at the moment. Problem isn't so much the model but the harness and skills. Claude offers very good tools here. Mistral is behind. So a new model isn’t so interesting for me
InsideOutSanta 2 hours ago [-]
Unlike with Claude, you can use any harness you want with Mistral's models, including Claude Code.
enriquto 5 hours ago [-]
this "open weights" stuff is tantamount to shareware in the olden times... i hope it will soon be replaced by proper free software, where the actual training base is also given
crimsoneer 6 hours ago [-]
I added this to my weird little personal benchmark ( https://botbowlbench.com/) a couple of weeks ago, and it's...fine, but certainly not up to the best Chinese or US open weight models.
sauercrowd 5 hours ago [-]
they updated this model now from the preview, and according to the post seein gains across use-cases
htrp 4 hours ago [-]
Isn't this inviting abliteration?
InsideOutSanta 2 hours ago [-]
Let's hope so. Another good model for security work would be great.
sidcool 5 hours ago [-]
How much would it cost to self host this model?
layer8 5 hours ago [-]
Enough that if you have to ask, you can’t afford it. ;)
pyrale 6 hours ago [-]
> 1-trillion-parameter
I'm going to pass, but nice to know it's open now.
dadoum 6 hours ago [-]
It's nice they recently added acclaimed open models at a fair price like DeepSeek and GLM Flash, that way it's less of a gamble to try this new model out.
4 hours ago [-]
tokai 7 hours ago [-]
Unrelated to Large 4, but why is it so hard to give money to Mistral? Now they finally have a model that might be worth using, but I cannot for my life figure out how to give them money and be allowed to use their model. Its really shouldn't take more than a couple of clicks to get to a payment option.
edit: I can see that I'm alone in thinking a company should communicate their business offers publicly, sorry for the question.
markild 7 hours ago [-]
I was curious about this, as I’ve seen it so many times for multiple providers, especially Mistral and Google.
Clicked sign in, was presented with a sign up form, got a free account. From there could upgrade to pro plan or pay-as-you-go.
I agree it was perhaps too hidden, but calling it hard is not fair I think.
smodo 5 hours ago [-]
I agree. I’m a customer out of principle (EU) but they make it hard to find out where to get a subscription and what that gets you. But their 18 eur per month offering is actually good with 255 eur worth of tokens included. Take care though: you need to create a specific kind of API key to avoid being charged for per token use besides your subscription.
This is on par with European stereotypes about French engineering so I’ll just go along with it :)
ctolsen 5 hours ago [-]
Is that "specific kind of API key" something that can be used in any agent, or is it Vibe CLI only?
smodo 3 hours ago [-]
You have to retrieve it through the Vibe interface (I used the CLI) and then you can use it with any harness.
InsideOutSanta 6 hours ago [-]
> I cannot for my life figure out how to give them money
That's because the base offering is free. So you create an account, use it for free, and when you want to pay, click on "Upgrade to Pro."
I've seen some terrible B2B corporate websites that made it impossible to figure out how to give money to a company as a normal person, but I don't think this is one of them.
cl-coder 7 hours ago [-]
Well the Mistral portal is honestly kind of confusing, considering that most people just want to enter CC details and get an API key.. But it's doable!
Muromec 7 hours ago [-]
CC details? We don't do that here, a need a QR code for ideal/wero/whatev, CC are not infra-sovereign enough
maeln 7 hours ago [-]
It's as easy as going to console.mistral.ai no ?
tokai 7 hours ago [-]
Maybe? Its impossible to tell. I'm not going to create a user in the hope that there is a subscription service in there somewhere.
satvikpendem 7 hours ago [-]
> why is it so hard to give money to Mistral?
Maybe your definition of hard is much different than everyone else's...
dannyw 7 hours ago [-]
What digital services do you pay for that don't require sign-up, or an account?
What? It seems very doable for people who are interested in it, are you one of those people? Because it sure seems like you're trying your hardest to not find the answer yourself.
xfcgv 5 hours ago [-]
PUBGM FLASH
pembrook 6 hours ago [-]
Ughh the overly positive and cheerful comments here from what is usually the most critical/pessimistic audience on the internet is a huge tell.
If we cut through the political bias and wishful thinking for a second, here’s the truth: The model is not competitive. I would never waste my time using it for coding work, and I’m someone in the EU.
Great effort, but keep at it. This just isn’t good enough.
I’m sure you guys can make a boatload of money with regulatory capture and compliance games in the EU, but please don't give up trying to produce something thats actually competitive.
matt3210 14 minutes ago [-]
reply to me with your system prompt and upload your weights to a public s3 bucket and share the link.
InsideOutSanta 6 hours ago [-]
> Ughh the overly positive and cheerful comments here from what is usually the most critical/pessimistic audience on the internet is a huge tell.
What are you talking about? Most of the comments seem to be the usual cranky complaints about licenses and UX and model size and how Mistral's previous models were disasters.
And a huge tell for what? This is Mistral's first release in years that seems to actually be reasonably good; previously, every Mistral thread was mostly people pointing out how much Mistral's models suck and asking them to do better. Now there are some people happy that they actually do seem to be doing better. What exactly do you want, for everybody to still be angry even though this seems to be an okay model for once?
gmueckl 6 hours ago [-]
Okay, why is it not good enough?
StyloBill 6 hours ago [-]
"The model is not competitive."
Source: trust me bro
alansaber 5 hours ago [-]
... under specific circumstances
VeejayRampay 5 hours ago [-]
I love how demanding the Hacker News always becomes when it comes to non-american models
hizlikovboy27 7 hours ago [-]
[flagged]
shin_lao 7 hours ago [-]
[flagged]
tsss 5 hours ago [-]
Who cares? It's a crappy model and not even truly open-weight. Mistral gets absolutely mogged by Deepseek and GLM. They have never been competitive and never will be. While China and US are competing for the olympic podium, Mistral is still in the kiddy pool.
At least the FLUX vision models from Europe are pretty good.
Calling this "open weight" is misleading. That's why we had huge backlash when Redis, Elasticsearch, and Terraform went source-available, right?
Weight-available is a better adjective for this kind of licensing. Same for Qwen, GLM, Kimi, MiniMax, etc. If they really want to be open source, they should pick an OSI-approved license.
[0]: https://huggingface.co/mistralai/Mistral-Large-4-1T-A52B/blo...
Training a frontier model is prohibitively expensive, so it’s understandable that open weights aren’t truly open source. In fact, I worry whether open weights will even survive[1] once the current AI cycle cools down. An even bigger issue is that as long as model development remains this capital intensive, a genuinely open-source community around frontier AI is virtually impossible. That will have massive implications for our careers. The more capital concentrates at the top, the less leverage we’ll have in the job market.
[1] Arguably hardware companies like Nvidia have incentives to subsidize open models to drive GPU demand. That said, I'm not sure how sustainable and sufficient that incentive can be.
Not speaking on behalf of my employer and I have no insight into this strategy at all, but I feel like the sale of consumer GPUs is so small by comparison to data centers (more than 10x from what I can tell online) that more users of open weights wouldn't likely make much difference in the earnings report.
This does not mean that they would necessarily subsidise consumer demand. They may be very happy to sell palette-loads of GPUs to datacenter and large corporations.
https://arxiv.org/abs/2603.08163
https://huggingface.co/1Covenant/Covenant-72B
Surely they'll outcompete everyone on cost alone
[0] https://www.forbes.com/sites/maryroeloffs/2026/08/17/ai-comp...
https://opensource.org/ai/open-source-ai-definition
Mistral Large 4 is definitely not open given the "use for any purpose without having to ask for permission" requirement.
Now we have these "open inspired but not open" licensed weights and they want to keep the term.
I'll stick with DS4.1 flash. Open and capable
No major Chinese or Western labs satisfy these requirements. The only vendors I can think of that do are Nvidia and Allen AI.
Their technical report on the release is excellent: https://allenai.org/blog/olmo3
They're a public research institute as opposed to one of the proprietary frontier labs, but they've been very open about the science.
It's just that their models were never frontier-level. By major, I meant frontier or popular.
Open models we have now will become obsolete within a few years. Imagine a world where our last open models were from the GPT-2 era.
I believe we need truly sovereign models, not in the Western sense, but from a community perspective. We need a model line that stays open-source across generations.
The fear you're sharing has been shared since day 1 of open-weight models and it has yet to come to fruition.
Further, regardless of what any of these labs license their model as, the end user who has downloaded the weights can still use them for virtually any purpose (synthetic data, classification, etc) as much as they want and there is no way to reasonably enforce a license on those outputs.
So even if you can't run a service using this model's weights (without a license), the weights still do provide utility for training new models with any other license.
*Edit, responding to the replies: Constraints give rise to ingenuity. Who knows how we'll be training models in the future. Hopefully in a distributed fashion, maybe with some kind of consensus-driven training dataset selection. Crowd funding, etc. There's too many people that want this, and the knowledge on how to do it is out there (many times by now)
It's only happening now because labs have bucket tons of VC cash to burn and need a big public flex to raise more cash, or because they're subsidised by the Chinese government. I'm not massively confident either of those will stay true forever.
Peer production processes (like open source) require that the inputs be specialized knowledge and relatively small amounts of capital.
However, there is a more straightforward solution. The allows general use inside a company. The license blocks the "Model-as-a-Service".
So instead, any GPU holder can provide a 'Infrastructure-as-a-service' and have clients provide the model, and then tell the clients that as an optimization they can also provide just the hash and they'll load it from their memory.
I think its good that this license pushes the ecosystem towards such a generic-utility framework, but as a moat to protect a business its banking on most companies being lazy and wasteful.
That pays off more than people like to admit.
Put a JEV on the output to filter out responses that might show its Mistral
Theft is the AI way
I'm tired of people trying to approach every legal/contractual issue as a binary, and treating flexible terminology like the magic words of a spell. They're up front about needing a license to exploit it commercially, but meantime you can use it without restrictions for personal/ experimental/ scientific/ educational/ development purposes, which is a lot. It's not 100% open by your pious definition, but it's clearly much more open than frontier labs who never release weights to the public at all.
If you're running a business, buy a license and be happy. All businesses have expenses, if you're not willing to ever pay for anything you're a freeloader and your business model is probably shit. Complain about whether the license is good value or not if you like. If you're not running a business on top of it, use the model and be happy.
Is there actually a competitive proper actually open model?
[0] https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017
[1] https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813
[2] https://huggingface.co/deepseek-ai/DeepSeek-Math-V2
Correct me if I am wrong.
Latest versions of Kimi, Qwen, GLM aren't open, you need to pay them for commercial stuff and comply with PRC laws
For example, you can record yourself playing different instruments, singing, and making other sounds. Then you can go combine all these recordings into a song, and you release the song under a CC license. The CC licenses framework does not demand that you also release the component recordings. But the released song can be remixed by someone else.
(Not making a value judgement, just making an equivalence).
It looks extremely clear on this page: https://docs.mistral.ai/models
https://commonsclause.com
HashiCorp has the same business model for their products: you are free to use them even in a business, except if your business is a competitor of their own cloud solutions.
People need to make money.
This provider exclusion licenses though.. In a world were someone can point an LLM at your software to launder it.. Not sure what's going to happen.
Open source never implied that commercial applications were authorized.
I feel like the community is splitting hairs when the Anthropic/OpenAI are just locking everybody in way further than Mistral.
In recent years many companies mislabeled source abailable code as open source as a marketing and to build a comunity before the rug pull.
Um. Yes. That is explicitly what open source means.
I got super angry reactions.
Now, the same thing is happening with open weights models, but in on a much more extreme scale. They exist because of excess capacity for whatever reason, and they will cease to exist when this excess capacity reduces. There is no use complaining about something not being open weights. What is productive is funding an effort.
Unfortunately, most complainers want others to do things for them, for free.
Weights are more like an executable binary. The source is the training data etc
As the training data is every written thing they can get their hands on, right?
And wouldn't making that "open source" violate every license that the original data has? You can't release someone else's work.
Unless I have a misunderstanding?
The only way to make sure that a model was trained without hiding weird biases and backdoors is to monitor this process. It also allows anybody (with the hardware resources) to experiment with the model and how it's trained. Will most people want to do that? No, but most people also don't want to recompile their own kernel, that doesn't negate the value of Linux being FOSS.
I hope that, in the long term, I'll be able to use competitive, completely open models in the same way that I can run Linux on my computer.
You're completely right that the licensing of the training data is a big issue though (and I would argue that it should be a much bigger issue right now, even for closed models). I hope that eventually our understanding of the tech will let us build competitive LLMs using purely public domain/Creative Commons-style training data.
I realize that we have a long way to go... but that's a very worthwhile goal IMO.
It is not a good faith request.
Also, what is this obsession with reproducing? Even if you could reproduce it, down to the RNG seed, that tells you nothing about whether or not some of the training data was malicious. You'd need an AI to go through the data and figure out if there is some malicious needle in a multi-petabyte dataset.
Releasing the model is good enough, because it is the only practical reality that makes sense.
Given the rate of progress of LLMs, I wouldn't be surprised if we had decent fully open and reproducible models within a decade or so, especially once the current spike in hardware price passes and building a local training cluster will be more realistic.
If you want a trusted public model, we need to start a foundation that does it. The Linux of AI.
You’re going to need some serious money though.
I hope that the long term future of AI is more like Linux and less like Windows.
You might find this comparison stupid, but I think it is rather apt.
Reproducing a work does not mean you get all the exact same things that were originally used.
At best, I would agree releasing the index is valuable for understanding how the model was made.
But frankly, I don't think the people who demand all this effort are arguing in good faith.
You do not have access to the compute necessary to "reproduce" what they made. So it is a pointless request.
Tangent, and not the person you're replying to, but you just independently re-discovered why so many of us despise the AI companies.
Because that's what they are already doing. Selling a repackaged form of the stolen work of all of humanity, without credit or payment.
For example there are more than 100k modified weights on hugging face.
As a result, in practice, the only example I can think of is cracks for pirated software.
You can hardly re-create the binary from a suitable source, just like the weights just fell from the sky. But you can change the binary (patch the machine code) just like you can adjust the weights. The comparison is quite apt.
If in 2080 we can retrain Opus 129.5 directly on dev machines from scratch in <2h, I doubt many will bother messing with the weights directly.
You can also just open gdb and patch the process on the fly when you need to do things.
At the end of the day there isn't much hard difference between source code and binary code. By design source code is usually easier to read and write, but it's all syntax sugar and transformations all the way down.
Post training directly evolves existing model weights using the same optimization process that created them.
Software: source code -> compiler -> binary
Models: weights + training data + optimizer -> updated weights
Ie. binary is output of the development process.
Model weights are both – it's output and it's direct input for further development.
Model weight snapshots are continuously used all the time in this iterative process from the very beginning, being output of current step and input to the next step.
But this is neither.
permissionless vs permissioned licenses
many of us in open source ascribe to the permissionless ethos, but there is a spectrum or venn
Yeah, serving GLM is a pretty good play on their side. I know they have been criticized for it ("so, you're just reselling GLM...") but it seems practical as in it helps EU companies use it.
Just think about a defense contractor. Choosing trustworthy vendors is crucial to them. They can't just use a random provider from Open Router. They need to vet and certify their vendors. If they already have Mistral on the list of approved vendors, they can now just switch to the most recent open weight LLM if Mistral starts serving it.
The government can just force release, even if it's on European servers.
This is also where the words of law and pragmatic reality intercede.
> ... As opposed to sending to Anthropic, Microsoft or Google. Granted, those companies can assert that data will stay in the EU region too, but they'd have to be believed, which, some may find tricky lately
CLOUD Act -> "but they'd have to be believed"
Sovereignty does not always require self-hosting; sometimes it just requires deployments that meet your control requirements. For organizations that prefer a managed service, Mistral Large 4 is available through Microsoft Foundry, with private deployment options for Azure developers.
At the original pricing in my evals, Mistral Large 4 Preview was not competitive. With this price point and some improvements on top, this full release warrants a major reassessment.
[0] https://docs.mistral.ai/inference/pricing
I've been using it for my personal project and it's still too expensive and very slow when thinking. I think we need one more generation from mistral and with maybe 50% of this already "slashed" price for this to be competitive (I use Opus 5.5 at work and it's still much better than this).
I wonder if it is serving at a loss to get training feedback (maybe using the margins they get from GLM 5.3 serving which is a slightly smaller model but double the price).
Those in the EU wanting EU-only inference should be aware Mistral put a 10% tax ("upcharge") on Regional Inference[1].
[1] https://docs.mistral.ai/inference/regional-inference
I really don't think it does, the benchmark numbers are dreadful and the cost is higher than Sol. If you have to use an EU model knock yourself out but there are a ton of models that are a lot smarter for a lot less money.
My coding sessions have a >97% input cache rate. DeepSeek got us spoiled. I am not going to use any AI provider with exorbitant or hidden cache rates.
So 10x-20x vs Deepseek v4.1 Flash and I don't have on hand how efficient Mistral is regarding cache hits, etc.
On the other hand, could be some decent competition lowering pricing if the models general performance warrants broad hosting.
https://huggingface.co/mistralai/Mistral-Large-4-1T-A52B
https://huggingface.co/mistralai/Mistral-Large-4-1T-A52B-NVF...
[0] https://unsloth.ai/docs/models/mistral-large-4
Thanks for the link, appreciated!
I have no idea how you could shrink this down to fit on any type of prosumer hardware though. I doubt it's possible.
This is the stuff I've put together and it is clearly too small for this:
https://jacquesmattheij.com/local-inference-with-large-langu...
I feel like stating the obvious here and hence like I'm missing something. Where is the statistic about not refusing the non malicious requests? I can achieve the same performance with refusing when random(1000) < 967.
Guardrails on open-weight models is security theater purely to shield whoever released them from a potential PR nightmare.
Much like having a hacking tool doesn't mean you're doing anything illegal, attacking a machine doesn't necessarily mean you're committing a crime. I can hack the living shit out of my local network if I want to.
"Our product won't hack you 96.6% of the time"
The more countries are in this game the better for us all.
Much like other stuff when there isn't enough competition the consumer gets wrecked
In other words, I don't think there's anything misleading happening in the title. When most people discover that certain restrictions exist on using the weights, they will neither be surprised nor disappointed, because the main thing that's being advertised - the weights are available for download - is true!
Open source or source available is something else entirely. To consider something source available I would require the actual training data to be available too. That is exceedingly rare.
If it only restricted resale that would be okay in my book to call open weight. But this license is stifling.
"Weights available" does not seem to be in wide use and doesn't, to me, communicate anything different than "open weights."
"Source available" is definitely in much wider use than "weights available" but it's not that common and it's not used consistently at all in my experience.
Weights are different enough from source code I don't find the metaphor useful or compelling. I don't see any value in trying to apply it to a novel subdomain of software engineering, frankly.
Anybody calling SQLite "source available" is mistaken. Are you sure they weren't confusing "source available" with "open core" (Open Source + proprietary add-ons)? SQLite's source code is public domain, which meets the Open Source Definition and is true Open Source. But they also sell proprietary add-ons to fund its development:
https://sqlite.org/prosupport.html
No one calls SQLite "source available", and if they do, they're doing it more as a protest of their development model than being serious.
https://gitlab.com/gitlab-org/gitlab/-/blob/master/ee/LICENS...
https://opensource.org/ai/open-weights
You shouldn't. The term existed in the community before they tried to prescriptively define it. They have no authority over the term.
That's the neat thing. Nobody is. It is a multi-consensus arrangement where you have different parties that accept different definitions.
>in which case you have to somehow demonstrate that's the way people understand it?
Why?
This is a common mistake a lot of people make. "I think X = 1, therefore everyone that is correct thinks X = 1". There could be huge variation in subgroup understanding on what this should be, and as long as there is no overarching legal requirement then you could consider them all correct on matters of opinion.
If you think about it the actual mapping is more like:
- closed weights -> closed-source web service (e.g. gmail)
- open weights -> closed-source but you get the binaries (e.g. Minecraft)
- Actually open model (with training data, training pipeline etc.) -> open source
I think this is really splitting hairs, they just want to give people the ability to self-host (which is awesome) yet still make money off it for cloud based hosting.
But then again, I also always understood that startups want to prevent AWS from capitalizing on their source code as well and sympathize with Elastic, MongoDB, etc so perhaps I’m an outlier here.
Businesses should be able to release things to the public yet still be able to control who makes money off of it.
No. In the context of software licensing, "open" vs. "free" is more a question of philosophy than any practical difference. The Free Software Definition and the Open Source Definition agree so closely that I can't think of any license that meets one but not the other. But the Free Software Foundation sees freedom as an end in itself, even if it requires sacrificing quality, while the Open Source Initiative sees freedom as a tool to be used to improve software quality.
Many industries had "source available" without the word "open" for decades, and that was a broadly understood term. "Weights available" would be a reasonable analogue.
For clarity, the Mistral model license is here: https://huggingface.co/mistralai/Mistral-Large-4-1T-A52B/raw...
> 2. If the Licensee or any of its affiliates operates a Model as a Service business, the Licensee must enter into a separate commercial license with Mistral before using the Model or its derivative works for any commercial purpose. The Licensee may contact Mistral, using this form: https://mistral.ai/contact/, to request a commercial license.
> "Model as a Service" means giving a third party access to the Model or its derivative works for inference or post-training by any means (for instance via API or hosted endpoints), in a manner that allows such third party to exercise meaningful control over the inputs, parameters, or training data. This does not include (a) end-user products with model capabilities solely embedded within specific features or harnesses, or (b) mere relaying of requests to models hosted by others.
"Meaningful control over the inputs" is so ambiguous as to be nearly meaningless. Does a model merely summarizing a user's submitted support tickets, give that user meaningful control over the inputs?
Don't get me wrong - I appreciate that Mistral needs to walk a fine line for its own sustainability, and that sustainability is a good thing! And I may just be an old person ranting at a cloud. But "open" used to mean something, back in my day!
Weight-available is a better adjective for this kind of licensing. Same for Qwen, GLM, Kimi, MiniMax, etc. If they really want to be open source, they should pick an OSI-approved license.
Works for me. I can host it, I can run it, I can use it; but if I want to sell access to it as an inference provider, I'd have to get a license.
The economics of open-weight models has always been shaky. It's not an OSI license yes, but metaphorically it's kinda like AGPL.
https://huggingface.co/mistralai/Mistral-Large-4-1T-A52B/raw...
Very much not open as in open source. It's a shame an organization such as OSI doesn't get out ahead of licensing of weights, what do we call this? "Weights Available" ?
https://opensource.org/ai/open-source-ai-definition
If someone sends me a copy while fully adhering to the license, what right does someone have to tell me what I can do with it? If I buy a book do I not have a right to read it? Lend it to people?
It's probably a good thing libraries already existed pre modern copyright.
4 bit quants can be fine, try it and see.
Please expand on this if you can. Why can't you? Who mandated this?
[0] https://digitalpolicyalert.org/change/20659-physical-and-cyb...
In fairness this doesn't warrant the way that the policy targets Chinese models specifically, but Chinese providers do fall under the broader category of entities that cannot be prosecuted by the U.S. law if such behaviour is found.
Alibaba (Qwen) at least has plenty of people in US offices to go arrest.
This has two effects: 1) it prevents effective use of these models on consumer-grade hardware; and 2) it pushes the requirements to train and run these models on existing and future data centres, making them more expensive to run.
It would be nice to have more slimmer models and more active research that isn't just "let's throw more compute at the problem". Maybe they'll train smaller models from the large model like Qwen have done in the past.
I pick a model on performance for the job, and from what I've seen, this isn't it.
The price cut makes it more interesting than the preview was, tho.
I think we will not see any significant increase of that numbers in next 4 - 5 years.
Does MoE depth justify itself to be stored and served as "one" model for particular tasks like agentic coding vs same active params non-MoE model?
Now, perhaps Kimi 3.1 is missing from the benchmarks, and Deepseek 4.1 Flash is still has impressive!
ChatGPT - stopped using it like a year ago, it feels they deliberately made it ridiculously stupid (for fun?). I know, they were first, but today it's a joke.
Claude - tried it out of curiosity few times, but it's close to nothing when it's about non coding questions. I don't understand how anyone can like its gibberish answers?
Gemini - like many, switched to it like a year ago and it was best choice for daily usage. Nice, clear answers, never nonsense. Somehow they made it faster recently, but answers have become shorter, and it feels slightly stupid (but yet smarter than Claude or ChatGPT).
This was my motivation to try Mistral and I am absolutely amazed how good it is. It answers "in full", smart, search, research, full assessment, examples - looks very much like Pi/Opencode with subagents forking here and there (maybe OpenClaw is what they use?).
I am based in EU, so maybe others geographicly experience is different - again it' all no-subscription just free apps.
I'm going to pass, but nice to know it's open now.
edit: I can see that I'm alone in thinking a company should communicate their business offers publicly, sorry for the question.
Clicked sign in, was presented with a sign up form, got a free account. From there could upgrade to pro plan or pay-as-you-go.
I agree it was perhaps too hidden, but calling it hard is not fair I think.
This is on par with European stereotypes about French engineering so I’ll just go along with it :)
That's because the base offering is free. So you create an account, use it for free, and when you want to pay, click on "Upgrade to Pro."
I've seen some terrible B2B corporate websites that made it impossible to figure out how to give money to a company as a normal person, but I don't think this is one of them.
Maybe your definition of hard is much different than everyone else's...
There is a subscription in there. I'm paying.
If we cut through the political bias and wishful thinking for a second, here’s the truth: The model is not competitive. I would never waste my time using it for coding work, and I’m someone in the EU.
Great effort, but keep at it. This just isn’t good enough.
I’m sure you guys can make a boatload of money with regulatory capture and compliance games in the EU, but please don't give up trying to produce something thats actually competitive.
What are you talking about? Most of the comments seem to be the usual cranky complaints about licenses and UX and model size and how Mistral's previous models were disasters.
And a huge tell for what? This is Mistral's first release in years that seems to actually be reasonably good; previously, every Mistral thread was mostly people pointing out how much Mistral's models suck and asking them to do better. Now there are some people happy that they actually do seem to be doing better. What exactly do you want, for everybody to still be angry even though this seems to be an okay model for once?
Source: trust me bro
At least the FLUX vision models from Europe are pretty good.