Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Qwen 3.8

twitter.com

Post reply on HN
Print view View original

361–370 of 793 posts

Re: Qwen 3.8

#361

Post by wood_spirit »

kelnos wrote: post #324
gardnr wrote: post #6

Earlier quoted context omitted.

It's hard to say what their motivation is. The Chinese firms seem to be working hard to commoditize intelligence which may be the most effective way to debase American frontier labs. And yeah: it also happens to be really good for humanity.

> It's hard to say what their motivation is. Feels pretty easy to me. They want to turn LLMs into a commodity, and watch the US AI labs crash and burn. There will still be plenty of customers who will pay them to host the models and run inference, even if the weights are open and others can offer competing products. (If necessary, the Chinese government can ban use of foreign inference services by Chinese citizens an…

As soon as the competition is bankrupted they no longer need to release for free? It’s like how big players enter markets by launching at a loss to destroy competitors?

Re: Qwen 3.8

#362

Post by Der_Einzige »

mceachen wrote: post #343
TheDong wrote: post #335

Earlier quoted context omitted.

I feel like there could also be a simpler explanation. Why does a debian contributor make debian free, why do they work on this thing anyone can use? Is it because linux and debian hate windows and iOS and want to see american fail? No, it's because most debian contributors believe software source code, information, should be free, users should be free to modify the code they use, and that they're building a thing th…

Please don't conflate a volunteer effort with no expected economic gain with a very well funded company (or fleet of companies). With the CCP's highly successful track record with subsuming other markets, Occam's razor applies to why they're doing this.

[flagged]

Re: Qwen 3.8

#363

Post by villish »

moffkalast wrote: post #345
villish wrote: post #339

Earlier quoted context omitted.

Do you have proof of that?

Do you have proof it's not? There are no laws they have to follow in regards to it, and it's practically impossible for them to go against their core self interest. More data is literally a direct component to better models and more revenue. If someone proves it's fake worst they get is a month of bad PR and then people will move on, as they always do.

> If someone proves it's fake worst they get is a month of bad PR and then people will move on.

You can say they stole from everyone to train their models in the first place and that's valid, but this isn't that. You are saying they are actively ex-filtrating data from any company using their services and lying about it.

Google/Apple/Microsoft or all of the dozen trillion dollar companies in the US would absolutely crush them in litigation. Neither OpenAI or Anthropic would be able to survive it. It's just not worth the risk.

Re: Qwen 3.8

#364

Post by vblanco »

storus wrote: post #58
SwellJoe wrote: post #34

Qwen is the most censored of the Chinese models in my testing, which makes me wonder in what other ways it is compromised. Open weights doesn't really reveal what's in there. And, in my tests, existing Qwen models are not at the pareto frontier of any metric; DeepSeek V4 Pro is better, faster, and much cheaper than Qwen 3.7 Max. (DeepSeek is also among the least censored of the Chinese models.) I guess we'll see if t…

DeepSeek V4 hallucinates like crazy and often forgets explicitly mentioned parts of the context. I guess compressing tokens and cherry-picking attention comes at a cost.

Deepseek V4 pro is a heavily undertrained model, they only trained it a bit more than the small version and that small version is 6-ish times smaller. Ive found that Flash is absolutely incredible as a workhorse for wide scale agentic nonsense, but Pro is a bit undercooked and really goes on strange tangents very often.

Re: Qwen 3.8

#365

Post by ahtihn »

fidotron wrote: post #209
baq wrote: post #78

Earlier quoted context omitted.

There’s a Twitter thread making rounds by Dean Ball about deceleration in AI development caused by open models and I can’t understand how people don’t see that it’s true: open models dismantle the frontier lab capex spend potential by reducing the training budget to zero in the limit. Tokens from different providers are not fungible, but customers are nevertheless very price sensitive and close enough is good enough,…

The big decelerationist threat is a sudden reduction in competition. If either OpenAI or Anthropic drop out or the open weights stuff is banned/becomes uncompetitive then the motivation and tolerance for taking risks with the larger training runs tanks. The closest we've seen to this in tech in recent decades was iOS vs Android, where Android only really was competitive for a very short window of time (approx 4.x) an…

> Android only really was competitive for a very short window of time

Complete non-sense. iOS and Android are equivalent. Users do not chose Android or iOS because one or the other is better.

It's just brand loyalty, status signalling and ecosystem lock-in that creates enough friction that people don't bother.

Re: Qwen 3.8

#366

Post by inigyou »

ahtihn wrote: post #365
fidotron wrote: post #209

Earlier quoted context omitted.

The big decelerationist threat is a sudden reduction in competition. If either OpenAI or Anthropic drop out or the open weights stuff is banned/becomes uncompetitive then the motivation and tolerance for taking risks with the larger training runs tanks. The closest we've seen to this in tech in recent decades was iOS vs Android, where Android only really was competitive for a very short window of time (approx 4.x) an…

> Android only really was competitive for a very short window of time Complete non-sense. iOS and Android are equivalent. Users do not chose Android or iOS because one or the other is better. It's just brand loyalty, status signalling and ecosystem lock-in that creates enough friction that people don't bother.

Only one of them lets you install unapproved apps.

Re: Qwen 3.8

#367

Post by glimshe »

derektank wrote: post #336
glimshe wrote: post #320

Earlier quoted context omitted.

"China" isn't giving anything. These are Chinese companies leveraging their best competitive strategy at the moment: competing on price.

Those Chinese companies are being funded by a substantial amount of government financing, I think at least 20% has come directly from state owned investment firms or government entities and probably more now. You obviously lose precision when you’re talking in sweeping terms like “China” but I don’t think it’s entirely unreasonable in this case. The Chinese government is playing a much more direct role in AI investme…

Nonetheless, the government isn't giving it away either. If China had a monopoly on the technology, or simply winning in quality, open weight models would never have seen the light of the day.

I just want to dispell the silly notion of altruism from China in this conversation.

Re: Qwen 3.8

#368

Post by a34729t »

baq wrote: post #78
gardnr wrote: post #6

Earlier quoted context omitted.

It's hard to say what their motivation is. The Chinese firms seem to be working hard to commoditize intelligence which may be the most effective way to debase American frontier labs. And yeah: it also happens to be really good for humanity.

There’s a Twitter thread making rounds by Dean Ball about deceleration in AI development caused by open models and I can’t understand how people don’t see that it’s true: open models dismantle the frontier lab capex spend potential by reducing the training budget to zero in the limit. Tokens from different providers are not fungible, but customers are nevertheless very price sensitive and close enough is good enough,…

I dunno, it means Anthropic and OpenAi need to get efficient and maybe cannot just expect trillion dollar ipos?

Re: Qwen 3.8

#369

Post by jnwatson »

vlovich123 wrote: post #348
mike_hearn wrote: post #311

Earlier quoted context omitted.

Really? The USA has built a ton of AI datacenters, exactly because it does have energy. The US IP system has flexed to allow training on all copyrighted content - compare that to Europe where such training is effectively forbidden. Britain doesn't even allow commercial web crawls! And the US has allowed the entire world to sign up and use its LLM APIs.

Consumer energy prices in China aren’t going up because of AI data centers. Easiest way to see they have an oversupply of energy, primarily due to solar.

That consumer energy prices are going up is simply a matter of public policy. Municipalities have the power to keep rates flat, but they choose not to.

Re: Qwen 3.8

#370

Post by tredre3 »

knollimar wrote: post #330
mike_hearn wrote: post #311

Earlier quoted context omitted.

Really? The USA has built a ton of AI datacenters, exactly because it does have energy. The US IP system has flexed to allow training on all copyrighted content - compare that to Europe where such training is effectively forbidden. Britain doesn't even allow commercial web crawls! And the US has allowed the entire world to sign up and use its LLM APIs.

Doesn't China smelt most of the world's aluminum due to inexpensive energy? I thought that was their thing

China smelts over 50% of the world supply. India/Russia/Canada does another 30%.

Your assessment is correct, those are countries with very low energy costs.

Post reply on HN