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

Kimi K3: Open Frontier Intelligence

kimi.com

Post reply on HN
Print view View original

441–450 of 1001 posts

Re: Kimi K3: Open Frontier Intelligence

#441

Post by tybit »

api wrote: post #415
lukebechtel wrote: post #386

> Chip Design > As an early proof of concept, Kimi K3 designed a chip to serve a nano model built on its own architecture. In a single 48-hour autonomous run, K3 built, optimized, and verified the chip using open-source EDA tools on the Nangate 45nm library. Within 4 mm², the chip closes timing at 100 MHz and sustains over 8,700 tokens/s decode throughput in simulation, packing 1.46M standard cells, 0.277 MB of SRAM,…

I had a thought a while back: sell large local models burned onto fused compute / ROM chips. Like cartridges for old game consoles. Slot (or probably plug into USB-C) and go. It’s an ASIC with the model wired into it so it’s very low power and fast. I’d buy these. Say $100 for a frontier class model. Maybe more.

Taalas is developing this, but not for Frontier class models. I hope that if we can least get the easy 80% of work done on that sort of hardware, we can greatly reduce the demand for GPUs, HBM and energy to some extent.

Re: Kimi K3: Open Frontier Intelligence

#442

Post by manacit »

Gecko4072 wrote: post #414

Looks like open models being months behind is a thing of the past. Now more like weeks.

Public disclosure of Mythos was April 7 and leaked happened in March, but it's been heavily delayed for well-known reasons.

That said, as the frontier moves, "months old" becomes more and more useful. Opus-tier models are being used to write serious software, so we're going to start seeing open models pick up a lot more usage imo.

Re: Kimi K3: Open Frontier Intelligence

#443

Post by hedora »

nikcub wrote: post #379
theplumber wrote: post #375

Earlier quoted context omitted.

[flagged]

> I pretty sure OpenAI and Anthropic are doing the same or worse. No they're not. It would end both companies if they were ever found to be doing that. Their terms are clear - if you use the coding plans they can[0] train in return. Enterprise and API, absolutely not. The argument here is that with the Chinese labs you have zero legal recourse. [0] opt-in, thanks

Anthropic constantly uses dark patterns to steal training data from customers (like the “how is claude doing” spam, data retention loosening when the safeguards false positive, etc).

Re: Kimi K3: Open Frontier Intelligence

#444

Post by smcleod »

wolttam wrote: post #104

I'm a bit nervous this one isn't going to be open-weights. Any mention of "open" has been struck from the literature for this model (it was present an hour ago). We don't even know active params? At this pricing, I'll be surprised if it's open.

It's 2.8T, I'm sure they will open the weights but it will only be able to be run on very high end machines.

Re: Kimi K3: Open Frontier Intelligence

#445

Post by ktzar »

TiredOfLife wrote: post #432
api wrote: post #415

Earlier quoted context omitted.

I had a thought a while back: sell large local models burned onto fused compute / ROM chips. Like cartridges for old game consoles. Slot (or probably plug into USB-C) and go. It’s an ASIC with the model wired into it so it’s very low power and fast. I’d buy these. Say $100 for a frontier class model. Maybe more.

Kinda already exists. demo https://chatjimmy.ai/ https://news.ycombinator.com/item?id=47103661

Hallucinates on the first question I ask, as 90% of these models that try to take shortcuts.

Re: Kimi K3: Open Frontier Intelligence

#446

Post by calgoo »

Cantinflas wrote: post #433
api wrote: post #415

Earlier quoted context omitted.

I had a thought a while back: sell large local models burned onto fused compute / ROM chips. Like cartridges for old game consoles. Slot (or probably plug into USB-C) and go. It’s an ASIC with the model wired into it so it’s very low power and fast. I’d buy these. Say $100 for a frontier class model. Maybe more.

You need terabytes of memory to run a frontier class model

I wonder, if you can run at 8k or 15k t/s, you could in theory run 10 or 20 agents (or more) at the same time and generate hundreds of versions, then just analyze them. Think thinking mode x1000 at least... Would be interesting to see how good it would be

Re: Kimi K3: Open Frontier Intelligence

#448

Post by ericd »

verdverm wrote: post #157
npn wrote: post #85

Not worth it. I have just tried a single prompt in the web interface and it is still not finish reasoning. It thinks too much and often repeats the same stuff over and over. Combine with the price it will surely more costly than gpt 5.6.

Its bad to judge these things on immediate release, there is a spike of excited users and that distorts performance. Also bad to judge from on a single interaction, you'll get bad requests with every provider, super busy times raise the probability

Yeah, I'm literally getting 529 API Overloaded responses on Claude Code right now.

Re: Kimi K3: Open Frontier Intelligence

#450

Post by fendy3002 »

api wrote: post #415
lukebechtel wrote: post #386

> Chip Design > As an early proof of concept, Kimi K3 designed a chip to serve a nano model built on its own architecture. In a single 48-hour autonomous run, K3 built, optimized, and verified the chip using open-source EDA tools on the Nangate 45nm library. Within 4 mm², the chip closes timing at 100 MHz and sustains over 8,700 tokens/s decode throughput in simulation, packing 1.46M standard cells, 0.277 MB of SRAM,…

I had a thought a while back: sell large local models burned onto fused compute / ROM chips. Like cartridges for old game consoles. Slot (or probably plug into USB-C) and go. It’s an ASIC with the model wired into it so it’s very low power and fast. I’d buy these. Say $100 for a frontier class model. Maybe more.

Your statement reminds me of Avenger's scene where Tony choosing Friday among other AI's catridges to use.

That sounds good and practical to happen!

Post reply on HN