5 ms·

No, they are not. They are winning because West is forbidden to use Chinese models for anything work-related.

by mordae 5mo ago

dyauspitr 5mo ago
Why? So that even more American IP can pass through Chinese servers? Or because their near frontier models are heavily government subsidized?
jxf 5mo ago
My understanding is that it's not that the _models_ are banned, but rather the _platform_ is banned. It is acceptable to host, say, `deepseek-r1-distill-qwen-7b` and run it yourself, for example. It is not acceptable (to the authors of these bans) to download the DeepSeek app and run it on your work device.
- forgotusername6 5mo ago
  We aren't allowed to use any unauthorized models even locally.
- eskibars 5mo ago
  I just left a job for a German B2B software company which sold primarily to large automotive, defense, and aerospace companies. Several of our customers specifically banned anything with the word "DeepSeek" -- hosted or self-hosted. There's still a lot of naivety on what the difference is between models and platforms, and its easier for a lot of these big companies to just make a blanket statement like "nothing DeepSeek" than for their procurement teams to try to understand and negotiate with each vendor. They don't see the potential benefit over the potential risk of somebody misinterpreting or getting it wrong, so they outright ban it. Most people that approve or buy software simply also just don't understand how models are being trained or if it's possible/how far a model could go to "introduce backdoors." A backdoor could be, from a business perspective, a model which has been trained to give answers that could hurt western business in a "strict text mode" or produces payloads in a programmatic mode that are intentionally trained to introduce software vulnerabilities. Anyone can make arguments against these for a variety of reasons (looking at the transparency of both sides and comparing, etc) but for many reasons today and for better or worse, many Chinese models are being banned on big software contracts, which gets back to the title of the article
  - wouldbecouldbe 5mo ago
    You are sending all your prompts code and files there. So ofcourse its an issue
    - overfeed 5mo ago
      Where's "there" on a self-hosted setup?
  - anvuong 5mo ago
    Thing is these models can also be a propaganda machine whether you run it locally or not. This is true no matter the origins. Chinese LLMs will never shit-talk CCP, and it will always give a rosy depiction of the Chinese government. It's perfectly understandable if companies don't want things like that. US/EU models have these problems too, but at least there are some ways to fight that: with a lawsuit or a megaphone on social networks. With Chinese models there is nothing you can do.
aspenmartin 5mo ago
You’re saying if we were allowed to use e.g. qwen more broadly the US wouldn’t be in the same strategic position? We have the best models…we own all the companies that make the best infra and the hyper scalers…I don’t think “oh we can use Qwen now?” Would exactly devastate the US
- visarga 5mo ago
  > I don’t think “oh we can use Qwen now?” Would exactly devastate the US You'd be surprised how useful it can be to fine tune it in enterprise.
  - aspenmartin 5mo ago
    Well definitely but we have plenty of sanctioned OSS options for that
- zozbot234 5mo ago
  Qwen's open models are quite small compared to Kimi, GLM and DeepSeek Pro, which are often described as near-SOTA.
pattt 5mo ago
Do we have any solid evidence these models can outperform Western models in terms of quality? Or is it more: because they are forbidden, they can't get enough training data, visibility etc. to compete?
- gpt5 5mo ago
  Scroll down to the leaderboard - https://arcprize.org/leaderboard https://arcprize.org/leaderboard Spoiler alert - they are all towards the bottom of the leaderboard. People come up with a wide variety of excuses for why they are not used despite being offered for significantly lower cost, but the answer is simply because they don't perform well enough for now.
  - aucisson_masque 5mo ago
    There isn't even deepseek V4. I'd rather trust LLM arena leaderboard, which puts it on par with sonnet.
    - gpt5 5mo ago
      LM Arena uses human side by side voting, which limits its applicability to complex tasks. The ARCPrize leaderboard does have Deepseek V3.2, which only scored 4% on ARC-AGI 2 (while the top models score over 80%). It also Kimi and Qwen, but they also didn't perform well.
thinkingtoilet 5mo ago
>No, they are not. They are winning You agree they are winning though, right? China is known for not playing fair, stealing industrial secrets, etc... that reputation matters and it's a good reason why the US is winning. Is the US perfect? No. Does the US play fair? No. Spare me the whataboutism in the comments. The bottom line is most people think the US is a safer bet and that's why we're winning. I personally wouldn't trust either government, but if I had to choose, I feel like I at least have a chance at secrecy and due process with the US. Obviously that is being eroded day by day, but you literally have no due process in China.
mariopt 5mo ago
True, many people don't know GLM 5.1 and Kimi 2.6, really on par with frontier models. There's also Minimax 2.7, DeepSeek 4, Qwen, Xiaomi 2.5 Pro, etc. China is leading in open source frontier models, so I don't really see how the US wins this one. At some point, companies and people will start running their own models in the cloud and locally, Chinese models will be everywhere.
- packetlost 5mo ago
  Nah, I model hop constantly as I work with serving GLM and Kimi models and they're not nearly as good as Opus 4.5+ and GPT 5.2+ and it's not particularly close. They're good by standards set a generation or two ago, but they're really not competitive with where the frontier models are at now.
  - zozbot234 5mo ago
    They compete with "mini" or "nano" model classes quite well given the price of inference. You'd need to "model hop" anyway, using Opus for everything is quite wasteful.
    - packetlost 5mo ago
      Now those aren't really "frontier models" now, are they.
      - zozbot234 5mo ago
        They are on the frontier of local models, where the game is often to get the best bang for the buck. You can always scale model size and compute (Mythos, GPT Pro, Gemini DeepThink) to reach better outcomes, but that's not a very interesting strategy.
        satvikpendem 5mo ago
        > They are on the frontier of local models That's not what anyone means when they say frontier models, don't change the definition. It's almost as bad as open weight being subsumed by open source when it comes to local models.
MetaWhirledPeas 5mo ago
> They are winning because West is forbidden to use Chinese models for anything work-related. Because the models hosted in China are not trusted. This is 100% a part of what makes up commercialization.
- aucisson_masque 5mo ago
  Deepseek is a fraction of the cost of western LLM and still just as good. I say it's also related.
- lmm 5mo ago
  Is anyone outside the US trusting anything hosted in today's US? If so, why?
  - coredev_ 5mo ago
    I would say that both US and China are using the data we trust upon them for industrial espionage. So don't use their models if you are working defence or other sensitive areas

No, they are not. They are winning because West is forbidden to use Chinese models for anything work-related.

by mordae 5mo ago

No, they are not. They are winning because West is forbidden to use Chinese models for anything work-related.

Why? So that even more American IP can pass through Chinese servers? Or because their near frontier models are heavily government subsidized?

jxf 5mo ago

My understanding is that it's not that the _models_ are banned, but rather the _platform_ is banned. It is acceptable to host, say, `deepseek-r1-distill-qwen-7b` and run it yourself, for example. It is not acceptable (to the authors of these bans) to download the DeepSeek app and run it on your work device.

forgotusername6 5mo ago
We aren't allowed to use any unauthorized models even locally.
eskibars 5mo ago
I just left a job for a German B2B software company which sold primarily to large automotive, defense, and aerospace companies. Several of our customers specifically banned anything with the word "DeepSeek" -- hosted or self-hosted. There's still a lot of naivety on what the difference is between models and platforms, and its easier for a lot of these big companies to just make a blanket statement like "nothing DeepSeek" than for their procurement teams to try to understand and negotiate with each vendor. They don't see the potential benefit over the potential risk of somebody misinterpreting or getting it wrong, so they outright ban it. Most people that approve or buy software simply also just don't understand how models are being trained or if it's possible/how far a model could go to "introduce backdoors." A backdoor could be, from a business perspective, a model which has been trained to give answers that could hurt western business in a "strict text mode" or produces payloads in a programmatic mode that are intentionally trained to introduce software vulnerabilities. Anyone can make arguments against these for a variety of reasons (looking at the transparency of both sides and comparing, etc) but for many reasons today and for better or worse, many Chinese models are being banned on big software contracts, which gets back to the title of the article
- wouldbecouldbe 5mo ago
  You are sending all your prompts code and files there. So ofcourse its an issue
  - overfeed 5mo ago
    Where's "there" on a self-hosted setup?
- anvuong 5mo ago
  Thing is these models can also be a propaganda machine whether you run it locally or not. This is true no matter the origins. Chinese LLMs will never shit-talk CCP, and it will always give a rosy depiction of the Chinese government. It's perfectly understandable if companies don't want things like that. US/EU models have these problems too, but at least there are some ways to fight that: with a lawsuit or a megaphone on social networks. With Chinese models there is nothing you can do.

aspenmartin 5mo ago

You’re saying if we were allowed to use e.g. qwen more broadly the US wouldn’t be in the same strategic position? We have the best models…we own all the companies that make the best infra and the hyper scalers…I don’t think “oh we can use Qwen now?” Would exactly devastate the US

visarga 5mo ago
> I don’t think “oh we can use Qwen now?” Would exactly devastate the US You'd be surprised how useful it can be to fine tune it in enterprise.
- aspenmartin 5mo ago
  Well definitely but we have plenty of sanctioned OSS options for that
zozbot234 5mo ago
Qwen's open models are quite small compared to Kimi, GLM and DeepSeek Pro, which are often described as near-SOTA.

pattt 5mo ago

Do we have any solid evidence these models can outperform Western models in terms of quality? Or is it more: because they are forbidden, they can't get enough training data, visibility etc. to compete?

gpt5 5mo ago
Scroll down to the leaderboard - https://arcprize.org/leaderboard https://arcprize.org/leaderboard Spoiler alert - they are all towards the bottom of the leaderboard. People come up with a wide variety of excuses for why they are not used despite being offered for significantly lower cost, but the answer is simply because they don't perform well enough for now.
- aucisson_masque 5mo ago
  There isn't even deepseek V4. I'd rather trust LLM arena leaderboard, which puts it on par with sonnet.
  - gpt5 5mo ago
    LM Arena uses human side by side voting, which limits its applicability to complex tasks. The ARCPrize leaderboard does have Deepseek V3.2, which only scored 4% on ARC-AGI 2 (while the top models score over 80%). It also Kimi and Qwen, but they also didn't perform well.

thinkingtoilet 5mo ago

>No, they are not. They are winning You agree they are winning though, right? China is known for not playing fair, stealing industrial secrets, etc... that reputation matters and it's a good reason why the US is winning. Is the US perfect? No. Does the US play fair? No. Spare me the whataboutism in the comments. The bottom line is most people think the US is a safer bet and that's why we're winning. I personally wouldn't trust either government, but if I had to choose, I feel like I at least have a chance at secrecy and due process with the US. Obviously that is being eroded day by day, but you literally have no due process in China.

mariopt 5mo ago

True, many people don't know GLM 5.1 and Kimi 2.6, really on par with frontier models. There's also Minimax 2.7, DeepSeek 4, Qwen, Xiaomi 2.5 Pro, etc. China is leading in open source frontier models, so I don't really see how the US wins this one. At some point, companies and people will start running their own models in the cloud and locally, Chinese models will be everywhere.

packetlost 5mo ago
Nah, I model hop constantly as I work with serving GLM and Kimi models and they're not nearly as good as Opus 4.5+ and GPT 5.2+ and it's not particularly close. They're good by standards set a generation or two ago, but they're really not competitive with where the frontier models are at now.
- zozbot234 5mo ago
  They compete with "mini" or "nano" model classes quite well given the price of inference. You'd need to "model hop" anyway, using Opus for everything is quite wasteful.
  - packetlost 5mo ago
    Now those aren't really "frontier models" now, are they.
    - zozbot234 5mo ago
      They are on the frontier of local models, where the game is often to get the best bang for the buck. You can always scale model size and compute (Mythos, GPT Pro, Gemini DeepThink) to reach better outcomes, but that's not a very interesting strategy.
      - satvikpendem 5mo ago
        > They are on the frontier of local models That's not what anyone means when they say frontier models, don't change the definition. It's almost as bad as open weight being subsumed by open source when it comes to local models.

MetaWhirledPeas 5mo ago

> They are winning because West is forbidden to use Chinese models for anything work-related. Because the models hosted in China are not trusted. This is 100% a part of what makes up commercialization.

aucisson_masque 5mo ago
Deepseek is a fraction of the cost of western LLM and still just as good. I say it's also related.
lmm 5mo ago
Is anyone outside the US trusting anything hosted in today's US? If so, why?
- coredev_ 5mo ago
  I would say that both US and China are using the data we trust upon them for industrial espionage. So don't use their models if you are working defence or other sensitive areas