The gap between open weights LLMs and closed source LLMs • Hacker News Board

Re: The gap between open weights LLMs and closed source LLMs

#151

Post by defrost » Sat, Jun 27, 2026, 6:05 AM UTC

cheesecakegood wrote: post #149

nmfisher wrote: post #143

Earlier quoted context omitted.

None of those companies are created by the Chinese government. They're obviously subject to the Chinese government, whose whims may change at any given moment, but as we're seeing at the moment, so are the American companies. And while I don't have a very positive view of the Chinese government, last I checked, they haven't been dropping bombs on innocent schoolchildren recently.

Bombs are a bit of a non sequitur here. The point is that Chinese companies are demonstrably hostile to American ones historically (and threatening in some specific structural ways to the American consumer). The presentation may be similar but to attribute American ethics to a Chinese decision is dubious.

Isn't the nature of capitalism such that many companies are demonstrably competitive (aka 'hostile' ?) with one another?

Chinese companies have also demonstrably pandered to the American consumer for many decades now.

To further muddy the waters, US companies have, some would argue, been openly hostile to the American consumer via monopoly practices, restricting access to purchased devices, etc.

Re: The gap between open weights LLMs and closed source LLMs

#152

Post by girvo » Sat, Jun 27, 2026, 6:09 AM UTC

jmyeet wrote: post #55

profsummergig wrote: post #5

IMHO, the biggest problem with the future of open weights models is that currently, open weights models are the result of philanthropy by some private org. (e.g. DeepSeek). The spigot can be turned off at any time. Until there's some sort of "community owned hardware", open weights models are always at risk of being discontinued.

How is this a complaint? Once you have the model, you have the model. Download DeepSeek-R1 671B and you have it. You might not get improvements in the future, just like you may not ever get a future release of an open source project. Is that an indictment of open source? But consider the alternative. OpenAI and Anthropic can shut off your account or API key at any time for any reason. How is this better? You have way…

> Download DeepSeek-R1 671B

Dunno why you'd want to though, considering v4 Pro (and even Flash) outpace it drastically

Re: The gap between open weights LLMs and closed source LLMs

#153

Post by girvo » Sat, Jun 27, 2026, 6:10 AM UTC

Schiendelman wrote: post #42

amunozo wrote: post #27

Earlier quoted context omitted.

Why are we assuming only American labs can innovate? DeepSeek already innovated a lot in efficiency, for example.

It's really unclear how much innovation DeepSeek has actually done, vs training on frontier model conversations.

Then you have no understanding of what DeepSeek has actually done. They publish their work openly: go have a look! Their architectural improvements are fascinating.

Re: The gap between open weights LLMs and closed source LLMs

#154

Post by knowaveragejoe » Sat, Jun 27, 2026, 6:17 AM UTC

ggm wrote: post #131

knowaveragejoe wrote: post #114

Earlier quoted context omitted.

I think the unspoken fear is that if we assume one or the other will "win" in reaching AGI(or whatever threshold of capability), the rest of the world will sooner or later live under their system of rule as a consequence

I very much doubt the primary reason nation states are lining up to permit or forbid access to these systems is 'fear of future AGI dominance' I think it's much more immediate/present: the weights and the information breach significant strategic controls on national data and posture, which can be back-derived from the models. If you can analyse a model, you can infer what structural inputs dictate it.

Can you expand on that reasoning more? Claude etc has national secrets hodden somewhere in its weights?

Re: The gap between open weights LLMs and closed source LLMs

#155

Post by echoangle » Sat, Jun 27, 2026, 6:34 AM UTC

Zetaphor wrote: post #121

echoangle wrote: post #82

Earlier quoted context omitted.

Believe me, if the government wants to stop you from having access to something like that, they could do it. Just give people some incentive to report you and make really harsh punishments and everyone will be thinking really hard about how bad they want have access.

Because that has worked so well for: * Drugs * Media piracy * Alcohol * Sex work * Unlicensed gambling The government is not an all powerful entity with absolute control over its people. Even in countries under past and present dictatorship there are examples of people getting access to what the government deemed as illegal.

I was thinking of this one:

https://en.wikipedia.org/wiki/Executive_Order_6102

Of course you’ll always be able to get access but the risk can be made so high that most people won’t try it.

There are countries that have death penalty on dealing with drugs and really severe prison terms just for having a small amount of drugs. There are still people that do it, but most people are effectively deterred because it’s just not worth it.

Re: The gap between open weights LLMs and closed source LLMs

#156

Post by 40four » Sat, Jun 27, 2026, 6:43 AM UTC

nmfisher wrote: post #143

40four wrote: post #125

Earlier quoted context omitted.

We should address the elephant in the room. The problem with the future of open weight models is not they are created as a result of philanthropy by some private org . All of the top contenders are created by the Chinese government . I don’t think we should describe these companies as simply releasing these highly capable open weight models out of the goodness of their hearts

None of those companies are created by the Chinese government. They're obviously subject to the Chinese government, whose whims may change at any given moment, but as we're seeing at the moment, so are the American companies. And while I don't have a very positive view of the Chinese government, last I checked, they haven't been dropping bombs on innocent schoolchildren recently.

Hey I hear you, I’m not trying to make this a political argument of who’s dropping bombs on who, or the American government is better than or worse than the Chinese. But what I said is a matter of fact.

We can debate the semantics of whether “created by” or “subject to” means the same thing in regards to the Chinese government, but that is neither here nor there.

I’m happy to take your wording that they are obviously “subject to” the Chinese government. That logically means they are subject to carrying out the CCP’s long term strategy. And as you said “whose whims may change at any given moment”.

That directly relates to the OP’s fears, that these models could be taken away at any given moment. “The spigot can be turned off at any time” as they put it.

Or another possibility is they will never turn the spigot off, but they will engineer it in a way to best achieve their goals. My bet is that’s the more likely outcome.

I simply disagree with the OP’s description of the problem as “open weights models are the result of philanthropy by some private org”, I think the problem is much more complicated than that

Re: The gap between open weights LLMs and closed source LLMs

#157

Post by disgruntledphd2 » Sat, Jun 27, 2026, 6:49 AM UTC

Onavo wrote: post #137

Eridrus wrote: post #136

Earlier quoted context omitted.

I think the bigger issue is the ever increasing capital requirements, which may cause even the closed weight companies to fall away from the frontier, e.g. Google & Meta are barely hanging on. For Google it feels a bit existential to remain at the frontier, but even then they're barely there. I hope that we find ways of continuing to improve these models besides continuing to exponentially increase capex spend until…

Google and Meta's failures are more due to mismanagement no?

At times of rapid change, having a working business model can be a disadvantage.

For instance, Facebook were able to optimize their core ads product for mobile, in a way that was much more difficult for Google.

Re: The gap between open weights LLMs and closed source LLMs

#158

Post by matheusmoreira » Sat, Jun 27, 2026, 6:50 AM UTC

profsummergig wrote: post #5

IMHO, the biggest problem with the future of open weights models is that currently, open weights models are the result of philanthropy by some private org. (e.g. DeepSeek). The spigot can be turned off at any time. Until there's some sort of "community owned hardware", open weights models are always at risk of being discontinued.

I wish we had some kind of distributed training capability... Like Folding@home, but for LLMs.

Re: The gap between open weights LLMs and closed source LLMs

#159

Post by woctordho » Sat, Jun 27, 2026, 6:51 AM UTC

matheusmoreira wrote: post #158

profsummergig wrote: post #5

IMHO, the biggest problem with the future of open weights models is that currently, open weights models are the result of philanthropy by some private org. (e.g. DeepSeek). The spigot can be turned off at any time. Until there's some sort of "community owned hardware", open weights models are always at risk of being discontinued.

I wish we had some kind of distributed training capability... Like Folding@home, but for LLMs.

See the recent advance of DiLoCo at Nous Research and Prime Intellect.

Re: The gap between open weights LLMs and closed source LLMs

#160

Post by woctordho » Sat, Jun 27, 2026, 6:54 AM UTC

echoangle wrote: post #82

anticorporate wrote: post #44

Earlier quoted context omitted.

There's also, importantly, a distinction between what are told we can no longer use, and what can actually be taken away. Open source and open hardware can be called illegal by a government, but, if we collectively invest our energy into open alternatives, they can't be taken away in the same sense. I can build a RepRap printer and I can use a local AI model. It's on all of us to make sure that the open alternatives…

Believe me, if the government wants to stop you from having access to something like that, they could do it. Just give people some incentive to report you and make really harsh punishments and everyone will be thinking really hard about how bad they want have access.

Fun fact: Hacker News is canonically banned in China, but I'm still talking here. There are plenty of techs to work around region block. The incentive to report somebody is comically called '50w' (500k CNY) and no one gives a shit about it in real life.