Kimi K3: Open Frontier Intelligence • Hacker News Board

Re: Kimi K3: Open Frontier Intelligence

#701

Post by fooker » Fri, Jul 17, 2026, 4:16 AM UTC

smith7018 wrote: post #366

eckr wrote: post #321

Earlier quoted context omitted.

These benchmark numbers are insane. The days when China was 6 months behind are over? How are they doing this with so much less resources than the US??? I have so much respect for the researchers there

I'm not sure where "so much less resources" comes from. Training the best model has nothing to do with having the most NVIDIA GPUs around. If that were true then xAI would have the best model. It comes down to the quality of data, research, and financial backing.

They have a pretty seamless workaround nowadays.

Build the data centers in middle eastern and other asian countries. Access it remotely.

There's no real way to restrict this without turning the internet into a draconian locked down shell of itself.

Re: Kimi K3: Open Frontier Intelligence

#702

Post by anigbrowl » Fri, Jul 17, 2026, 4:19 AM UTC

This might be the most impressive website generator demo I've seen: https://macos27.kimi.page

Context from the person who prompted it: https://x.com/mweinbach/status/2077827886149439547

Re: Kimi K3: Open Frontier Intelligence

#703

Post by zarzavat » Fri, Jul 17, 2026, 4:25 AM UTC

scrollop wrote: post #361

zarzavat wrote: post #243

Earlier quoted context omitted.

It's like reading Anthropic's obituary.

Nah: https://www.youtube.com/watch?v=LSlV206xPqM These real world examples show it's one tier away.

If Chinese AI companies can train a model that's slightly worse than the frontier, then there's no reason why they can't train a model that is slightly better than the frontier.

Everybody can agree that K3 doesn't clearly surpass Fable. However, inevitably there will be a time in the future when a Chinese AI company releases a model that's better than any US model.

K3 isn't the knockout blow but it's the 2nd knockdown that makes everyone in the arena realize that the fighter is not winning the fight.

Re: Kimi K3: Open Frontier Intelligence

#704

Post by joe_the_user » Fri, Jul 17, 2026, 4:27 AM UTC

rowanG077 wrote: post #418

vonneumannstan wrote: post #334

Earlier quoted context omitted.

[flagged]

Distillation is not an attack. It simply a way to train a model. Not doing it when you are behind is akin to snatching defeat from the jaws of victory.

Indeed,

Thinking about, if I had a lab, I'd be trying to get training data from a combination of the open web, piracy and all of my revivals. I wonder how labs are doing that.

Re: Kimi K3: Open Frontier Intelligence

#705

Post by qq66 » Fri, Jul 17, 2026, 4:30 AM UTC

Abishek_Muthian wrote: post #698

simonw wrote: post #111

Pelican: https://tools.simonwillison.net/markdown-svg-renderer#url=ht... - rendered via the OpenRouter API: https://openrouter.ai/moonshotai/kimi-k3 95 input, 16,658 output = 25 cents! https://www.llm-prices.com/#it=95&ot=16658&ic=3&oc=15 (13,241 of those were reasoning tokens.) I think that's the most expensive pelican I've rendered through a Chinese model so far.

For me none of the models have managed to accurately replicate any given .png/.jpg image in SVG. I guess that requires both vision encoders and coding layers to work perfectly.

have you tried quiver.ai

Re: Kimi K3: Open Frontier Intelligence

#706

Post by m00dy » Fri, Jul 17, 2026, 4:33 AM UTC

This will put Kimi's valuation at least $0.5T

Re: Kimi K3: Open Frontier Intelligence

#707

Post by braedon-dev » Fri, Jul 17, 2026, 4:33 AM UTC

lucianmarin wrote: post #683

boredatoms wrote: post #655

Earlier quoted context omitted.

I worry that some model provider will go and hire artists to draw pictures of pelicans on bicycles to make training data

Shouldn't a LLM know how a pelican looks like from biology without being trained on very specific artistic data?

The model likely does know what a pelican looks like; the hard part is translating that into ordered SVG path commands. It's really testing spatial and vector composition, not biological recall.

Re: Kimi K3: Open Frontier Intelligence

#708

Post by realusername » Fri, Jul 17, 2026, 4:46 AM UTC

anon373839 wrote: post #687

vl wrote: post #581

Earlier quoted context omitted.

This is strategy by Chinese government, so much of US economy is invested in AI. Releasing free or cheap versions of the models undermines US economic growth. It’s asymmetric strategy that makes sense if you are close second in AI race. If situation is reversed, US would do the same.

> so much of US economy is invested in AI. This is a talking point that plays into the frontier labs' desire to be seen as "too big to fail". While yes, several hundred billion dollars have been invested in AI, (a) much of this is in the form of circular Monopoly-money deals, and (b) the US GDP is over $30 trillion annually. The real economy - the one that makes food, builds homes, provides medical care, etc. - is so…

> The real economy - the one that makes food, builds homes, provides medical care, etc. - is so much bigger and more important than the AI industry.

The real economy has seen pretty poor growth under Trump tariffs chaos though and I'm not sure the US economy could survive a crash of the tech companies

Re: Kimi K3: Open Frontier Intelligence

#709

Post by g42gregory » Fri, Jul 17, 2026, 4:49 AM UTC

specialist wrote: post #528

softwaredoug wrote: post #478

So Chinese labs are driving essentially towards commodotized intelligence. Even if its a few months behind the US. Is this a classic 'commoditize my compliment' situation? They want to sell the hardware and infrastructure behind AI and make the software part not the value driver / moat? I can see it. But also even two Chinese labs sinking 100s of millions USD into training isn't exactly commoditization. It's still a…

It's an attack on USA's AI tech giants. Destroy any hope of profitability, prevent further capitalization, ultimately bankrupt them. Hasten the popping of the AI bubble. Drag down the stock market. Put a dent in the GDP (alleged growth). Cause investors to pull back, further depressing economy. Devalue the dollar. Kicking off an interest rate doom loop. Challenge USA's hegemonic role in the emergent multi-polar (post…

US "AI tech giants" are foaming at the mouth about "dangers" ans "safety", about "how US customers can not be trusted with AI", how much everybody should pay them, and how many $Trillions should they make, at US customers' expense.

For each "AI tech giant", there are millions of other US companies that will benefit from broader AI availability. US economy (and US tech economy) != Anthropic + OpenAI.

Re: Kimi K3: Open Frontier Intelligence

#710

Post by SXX » Fri, Jul 17, 2026, 5:02 AM UTC

wilg wrote: post #629

zwu-var wrote: post #598

Earlier quoted context omitted.

> Companies can still make money from commodities Especially Chinese companies. Just think about all the other industries where Chinese companies dominate by extremely low cost.

Isn't that typically because they have lower labor costs? Not sure that applies as much to AI.

To be honest I do not thihk you are wrong. Not in a way AI talent is underpaid, but they unlikely need to offer $1,000,000 sign up bonuses to just create AI lab. Its still pays better than average SWE job, but not x5-x10 more like in US.

Though in manufacturing China is dominating because of immense optimized supply chain, economy of scale and now workforce with know-how.

I honestly cant compare with US since I'm not from US, but to produce something in EU you will have weeks or months lags on logistics of source materials and components. And some industries only exist in some countries, etc.

In China in one single place you have anything you need to produce from a toaster to a car in hands reach or might be few days at most.

And I guess China did the same with education. Education is a lot cheaper there so they just have more talent.