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

GLM 5.2 Is Out

twitter.com

Post reply on HN
Print view View original

381–390 of 544 posts

Re: GLM 5.2 Is Out

#381

Post by scotty79 »

Havoc wrote: post #202
khalic wrote: post #61

Given the US government’s latest stunt with Fable, this is looking more and more like the future. Can’t rely on strategic products if they’re gated by capricious actors. Open weight models are basically immune to that

It’s very likely the Chinese go dark too the second they have parity / lead

I wouldn't bet on it. Chinese live the free market ideals instead of just preaching them but rent-seeking and seeking regulatory capture at the first opportunity. In China business doesn't control politics. Dynamics is completely different and so might be the outcomes.

Re: GLM 5.2 Is Out

#383

Post by kouteiheika »

ddxv wrote: post #238
thewebguyd wrote: post #87

Earlier quoted context omitted.

> Open weight models are basically immune to that Somewhat. The US Gov can make it illegal to transact with, download, use, etc. foreign open weight models. Of course, enforcement will be difficult for individuals (businesses will comply by default, and they would all be pulled off Github and other US based hosting locations if they went the sanctions route). But, we are also quickly going down the road of frightenin…

I think that this is what OpenAI/Anthropic want but they wont say it publicly. The will be OK with the US banning regulating and banning open source models as it let's Anthropic and OpenAI charge huge premiums to American business clients for their models. Also the marketing of them getting to say "our models are so dangerous" only a few companies or select users are allowed to use (benchmark) them would help keep th…

> I think that this is what OpenAI/Anthropic want but they wont say it publicly.

Won't say it publicly? Anthropic is openly and explicitly saying it publicly. Here: https://darioamodei.com/post/policy-on-the-ai-exponential

> AI companies that develop advanced AI models must have strong security standards that protect their model weights

If the model is open-weight then there's nothing to protect, so the only way to fulfill this requirement is to make open-weights illegal.

Re: GLM 5.2 Is Out

#384

Post by blurbleblurble »

treefry wrote: post #298
bxclltkfz wrote: post #98

Earlier quoted context omitted.

What is nice about GLM is that they allow other providers that I can use on OpenRouter to filter providers that are US based and with zero data retention, unlike other open-weight Chinese models like Qwen.

Unless you self host, zero data retention cannot be guaranteed.

Confidential inference gets us closer (via Nvidia's TEE architecture)

Re: GLM 5.2 Is Out

#385

Post by MiscIdeaMaker99 »

abc42 wrote: post #369

Genuine question: How safe is it to use Chinese models via their services? Surely Anthropic and OpenAI are ingesting what I push there as well, but they're at least vaguely allied with my home country geopolitically. China on the other hand seems to be interested in supporting countries like Iran and Russia.

I don't know what you consider to be safe, but I assume you're talking about privacy and data retention.

You can pay someone like Ollama, which is US-based, which runs those models in their cloud. You can also download many/most of the models to run them locally on your own computer, although the performance depends entirely on your hardware capabilities.

Ollama has their own policies about privacy and data retention here:

https://ollama.com/privacy

That's just one third-party doing this, and there are many others.

Re: GLM 5.2 Is Out

#386

Post by victorbjorklund »

khalic wrote: post #345
Reubend wrote: post #38

Seems like there's no official blog post with benchmark results yet. But I'm once again thankful for the Chinese AI labs for being open with their work and contributing it to the world under permissive licenses like this. The Fable 5 fiasco is just another reminder of how valuable these things are to have.

It's also a reminder that as soon as Chinese models take the lead, they will switch to closed source too... so let's not be complacent, we need stronger, completely open data models, open source code, etc. to mitigate this risk

Based on what? Do you have real proof on it or is it just a guess that Chinese companies aren’t better than American ones?

Re: GLM 5.2 Is Out

#387

Post by vidarh »

amosjyng wrote: post #358
vidarh wrote: post #331

Earlier quoted context omitted.

Harness certainly matters a lot, though GLM is pretty forgiving. I just had Opus tell me that based on numbers over the last week, from quite a few billion tokens total across half a dozen providers, GLM 5. 1 has been more reliable for one of my projects than Sonnet... Just switching on 5.2 now.

How are you collecting your metrics on token usage and reliability?

They are from my own runs, with reliability measured in terms of passing extensive test suites. So caveat is that this applies for my specific use and might well vary greatly.

Re: GLM 5.2 Is Out

#388

Post by hadlock »

Alifatisk wrote: post #368
re-thc wrote: post #346

Earlier quoted context omitted.

> That's because Qwen's flagship models are not, in fact, open weight They changed course when they fired the old lead and hired a new 1 from ex-gemini.

No, Qwen Max series has always been proprietary.

They also stopped releasing 100b+ model weights after firing him

Re: GLM 5.2 Is Out

#389

Post by LogicFailsMe »

khalic wrote: post #361
cududa wrote: post #360

Earlier quoted context omitted.

How do you figure that? “also a reminder that as soon as Chinese models take the lead, they will switch to closed source too” What specifically about their release strategy “reminded” you of that conjecture? The premise that they only open source the models … because it somehow helps them leapfrog American labs, and once they actually can leapfrog them, they’d close source them, doesn’t really track for me. Am I miss…

The point I’m making is that this has become a strategic resource. The Chinese government allows wide sharing of their models because is weakens the US position. If Chinese models become better than Americans, do you believe the CCP will allow the free distribution of their flagship models? Think again if it’s the case.

They would still be at a significant compute disadvantage and deploying them worldwide seems to be how they work around that currently as they put together a homegrown alternative.

Re: GLM 5.2 Is Out

#390

Post by LogicFailsMe »

tshaddox wrote: post #263
b3ing wrote: post #190

Earlier quoted context omitted.

Just like we can’t allow Chinese EVs in the USA, because we can’t and don’t want to compete. VPN usage would go up, to get the banned models.

In the United States it’s illegal to sell Chinese EVs. It’s also illegal to download copyrighted music and movies. Which one do you suppose illegal open-weight models would more closely resemble?

Because they're doing such a gangbusters job here keeping copyrighted music and movies from being download?
Post reply on HN