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

Kimi Released Kimi K2.5, Open-Source Visual SOTA-Agentic Model

kimi.com

Post reply on HN
Print view View original

201–210 of 251 posts

Re: Kimi Released Kimi K2.5, Open-Source Visual SOTA-Agentic Model

#201

Post by side_up_down »

johndough wrote: post #81
endymi0n wrote: post #45

Earlier quoted context omitted.

One. Trillion. Even on native int4 that’s… half a terabyte of vram?! Technical awe at this marvel aside that cracks the 50th percentile of HLE, the snarky part of me says there’s only half the danger in giving something away nobody can run at home anyway…

The model absolutely can be run at home. There even is a big community around running large models locally: https://www.reddit.com/r/LocalLLaMA/ The cheapest way is to stream it from a fast SSD, but it will be quite slow (one token every few seconds). The next step up is an old server with lots of RAM and many memory channels with maybe a GPU thrown in for faster prompt processing (low two digits tokens/second). At t…

I'd take "running at home" to mean running on reasonably available consumer hardware, which your setup is not. You can obviously build custom, but who's actually going to do that? OP's point is valid

Re: Kimi Released Kimi K2.5, Open-Source Visual SOTA-Agentic Model

#202

Post by storystarling »

wongarsu wrote: post #78
endymi0n wrote: post #45

Earlier quoted context omitted.

One. Trillion. Even on native int4 that’s… half a terabyte of vram?! Technical awe at this marvel aside that cracks the 50th percentile of HLE, the snarky part of me says there’s only half the danger in giving something away nobody can run at home anyway…

Which conveniently fits on one 8xH100 machine. With 100-200 GB left over for overhead, kv-cache, etc.

The unit economics seem pretty rough though. You're locking up 8xH100s for the compute of ~32B active parameters. I guess memory is the bottleneck but hard to see how the margins work on that.

Re: Kimi Released Kimi K2.5, Open-Source Visual SOTA-Agentic Model

#203

Post by hex4def6 »

esafak wrote: post #143
PlatoIsADisease wrote: post #130

Earlier quoted context omitted.

>The model absolutely can be run at home. There is a huge difference between "look I got it to answer the prompt: '1+1='" and actually using it for anything of value. I remember early on people bought Macs (or some marketing team was shoveling it), and proposing people could reasonably run the 70B+ models on it. They were talking about 'look it gave an answer', not 'look this is useful'. While it was a bit obvious th…

Hear, hear. Even if the model fits, a few tokens per second make no sense. Time is money too.

If I can start an agent and be able to walk away for 8 hours, and be confident it's 'smart' enough to complete a task unattended, that's still useful.

At 3 tk/s, that's still 100-150 pages of a book, give or take.

Re: Kimi Released Kimi K2.5, Open-Source Visual SOTA-Agentic Model

#204

Post by stopachka »

Is there a startup that takes models like this, and effectively gives you a secure setup, where you have (a) a mobile app that (b) talks to some giant machine that only you have access too.

If a 10K computer could run this, it may be worth it to have a "fully on prem" version of ChatGPT running for you.

Re: Kimi Released Kimi K2.5, Open-Source Visual SOTA-Agentic Model

#205

Post by darrinm »

storystarling wrote: post #197
zozbot234 wrote: post #163

Earlier quoted context omitted.

"tool call" is just a reference to any elementary interaction with the outside system. It's not calling third-party APIs or anything like that.

True, but that's still 1,500 inference cycles. Even without external API fees, the latency and compute burden seems huge. I don't see how the economics work there without significant subsidies.

FWIW many tool calls can be and often are made in one inference cycle.

Re: Kimi Released Kimi K2.5, Open-Source Visual SOTA-Agentic Model

#206

Post by conception »

DJBunnies wrote: post #127
conception wrote: post #123

Earlier quoted context omitted.

https://clocks.brianmoore.com K2 is one of the only models to nail the clock face test as well. It’s a great model.

Cool comparison, but none of them get both the face and the time correct when I look at it.

Refresh. It’s not every time but k2 hits a perfect clock for me about 7/10 or so.

Re: Kimi Released Kimi K2.5, Open-Source Visual SOTA-Agentic Model

#207

Post by in-silico »

Alifatisk wrote: post #93
Reubend wrote: post #13

I've read several people say that Kimi K2 has a better "emotional intelligence" than other models. I'll be interested to see whether K2.5 continues or even improves on that.

Yup, I experience the same. I don't know what they do to achieve this but it gives them this edge, really curious to learn more about what makes it so good at it.

A lot of people point to the Muon optimizer that Moonshot (the creators of Kimi) pioneered. Compared to the standard optimizer AdamW, Muon amplifies low-magnitude gradient directions which makes the model learn faster (and maybe gives Kimi its unique qualities).

Muon paper: https://arxiv.org/abs/2502.16982

Re: Kimi Released Kimi K2.5, Open-Source Visual SOTA-Agentic Model

#208

Post by culi »

motoboi wrote: post #80
bertili wrote: post #34

The "Deepseek moment" is just one year ago today! Coincidence or not, let's just marvel for a second over this amount of magic/technology that's being given away for free... and how liberating and different this is than OpenAI and others that were closed to "protect us all".

What amazes me is why would someone spend millions to train this model and give it away for free. What is the business here?

All economically transformative technologies have done similar. If it's privatized, it's not gonna be transformative across the industry. The GPS, the internet, touchscreens, AI voice assistants, microchips, LCDs, etc were all publicly funded (or made by Bell Labs which had a state-mandated monopoly that forced them to open up their patents).

The economist Mariana Mazzucato wrote a great book about this called The Entrepreneurial State: Debunking Public vs. Private Sector Myths

Re: Kimi Released Kimi K2.5, Open-Source Visual SOTA-Agentic Model

#209

Post by culi »

mekpro wrote: post #139
PlatoIsADisease wrote: post #132

Earlier quoted context omitted.

I am convinced that was mostly just marketing. No one uses deepseek as far as I can tell. People are not running it locally. People choose GPT/Gemini/Claude/Grok if you are giving your data away anyway. My biggest source of my conspiracy is that I made a reddit thread asking a question: "Why all the deepseek hype" or something like that. And to this day, I get odd, 'pro deepseek' comments from accounts only used ever…

Except that, In OpenRouter, Deepseek always maintain in Top 10 Ranking. Although I did not use it personally, i believe that their main advantage over other model is price/performance.

Fifth in market share in fact!

https://openrouter.ai/rankings

There are a lot of applications where you really just want a cheap and efficient model that's still somewhat competitive and that's exactly the niche DeepSeek fulfills the best.

Re: Kimi Released Kimi K2.5, Open-Source Visual SOTA-Agentic Model

#210

Post by culi »

conception wrote: post #123
Alifatisk wrote: post #53

Have you all noted that the latest releases (Qwen3 max thinking, now Kimi k2.5) from Chinese companies are benching against Claude opus now and not Sonnet? They are truly catching up, almost at the same pace?

https://clocks.brianmoore.com K2 is one of the only models to nail the clock face test as well. It’s a great model.

Kimi 2 is remarkably consistently the best. I wonder if it's somehow been trained specifically on tasks like these. It seems too consistent to be coincidence

Also shocking is how the most common runner up I've seen is DeepSeek

Post reply on HN