Why are there so many pictures of Sam Altman just Looking Like That?
If I were him I too would be in disbelief that people are buying my bullshit 24/7. What other expression could express this so well?
Just a day ago, a senior Anthropic executive claimed the U.S. still held a 6 to 9 month lead in frontier AI models, while calling Chinese model distillation adversarial. One day later, Moonshot’s Kimi K3 beat Claude Fable 5 on Frontend Code Arena. And the funniest part is that it is an open weight model.
So the whole 6 to9 month lead lasted about 24 hours. 🤣
https://hai.stanford.edu/news/inside-the-ai-index-12-takeaways-from-the-2026-report
That investment graph is insane
Totally not a bubble
Anything is possible with a sprinkle of fraud!
Tbf they (Moonshot) claim it’s not as good as Fable overall. Fable is also held back by US government mandated guards. Anthropic could well be ahead by a lot in terms of research and maybe we just don’t have the full picture. But that puts them in an even worse spot because it would mean they’re already at the limit of what they can compete with in the market.
I think the big picture here is that the difference in quality is largely subjective at this point, while US companies are burning through orders of magnitude of cash which is obviously not sustainable. We shouldn’t underestimate the power of developing things in the open. Chinese open models benefit from the wisdom of an entire global research community while American engineers working on proprietary closed models are working in their own insular silos. It should be no surprise that the scientific community at large would pull ahead of these small teams. On top of that, doing research in the open amortizes the cost. Incidentally, this is exactly the same logic that led open source to dominate in recent years.
It doesn’t hurt that American frontier models are phenomenally huge and cost an exorbitant amount per token in and out while the main source of real world value created by llms in the past year has been through agentic/harness/other words for massive context systems that show a real benefit from simply using more tokens.
Kimi K3 looks insane but it’s also very expensive right now, hopefully that changes or I don’t see myself using it over the previous models
I thought Anthropic was heavily subsidizing their models but KimmyK is currently cheaper than Fable and is allegedly a pretty good deal. The only thing it’s missing is a comprehensive coding environment like Claude Code and the whole MCP and plugins ecosystem.
If Anthropic pulls Fable then a Chinese lab will be in front of them for the first time I can think of.
Kimi K3 works through OpenCode/Pi. It just got added to OpenCode Go but it’s practically unusable due to the cost (but they might negotiate it down soon), right now you have to use it through OpenRouter or official API, or Kimi Code subscription
I don’t get this. How can Chinese models be so performant if they are spending a tiny fraction of US investment. If this difference in investment is down to research investment, then there is something I’m not getting? And, if Chinese models are open, how is it that they can keep ahead of US? Finally, if all this is true, the investors in US AI business is dangerous territory … i would be running away from that inflated investment. Of course i barely understand AI …
Very cheap electricity + a stronger education system + having access to the work of other open-source groups
Partly, the Chinese are distilling closed source models, and OpenAI and Anthropic’s attempts to stop this have only succeeded in making their outputs less transparent. Partly, the actual frontier of AI is currently mostly vibes and marketing, as everyone’s cheated the tests so badly that the only way to stay in the running is to cheat. Partly, that weve passed the point where even an honest evaluator could say which general purpose model is better than another, since their performance fluxuates wildly from task to task.
*fluctuates
Chinese schools do have reputation for being extremely difficult and China invests more in actual education.
Combine that with most US funding being a giant Ourobouros of AI companies funding eachother…
I mean if it’s going against Anthropic pricing that’s steal a good deal.
First bomb they’ve dropped in 50 years, wow.
check notes
Uhhhh not sure about bombing, but they went straight to clubbing people to death.
https://en.wikipedia.org/wiki/2020–2021_China–India_skirmishes
Lmao, how many people died? Honestly?
Aren’t you nonconfrontational?
Is clubbing people to death something you excuse, but only for China!?
How many people were killed in this brutal war that must have left countless dead in its wake?
A single death is too much, unless it is China.
What a goober.
Lmao, how many???
Stop pretending you give a duck.
The latest model is K2.7, I can decide what wrote the article. Is it more likely than an Ai did not know 2.7 existed or a journalist at Gizmodo is incompetent?
What do you mean by this? The latest model is K3, which was released yesterday.
Edit: ah I see the reference to K2.6
I did not know K3 was released until I started working today.
Ah gotcha, well you hit a good point anyway as they mention the prior latest model as K2.6, which does seem like a glaring error. K2.7 has been out for a month!
yay
