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

Kimi K2.6: Advancing open-source coding

kimi.com

Post reply on HN
Print view View original

241–250 of 394 posts

Re: Kimi K2.6: Advancing open-source coding

#241

Post by disiplus »

simonw wrote: post #53

Accessed via OpenRouter, this one decided to wrap the SVG pelican in HTML with controls for the animation speed: https://gisthost.github.io/?ecaad98efe0f747e27bc0e0ebc669e94... Transcript and HTML here: https://gist.github.com/simonw/ecaad98efe0f747e27bc0e0ebc669...

was part of the beta, its properly good model, in some sense i forgot that im not on opus or gpt. opus is still better. gpt is the one struggling for me. it has some niche in backend work but you can get the same with opus with skills, its lacking in almost all others.

Re: Kimi K2.6: Advancing open-source coding

#242

Post by darksaints »

codemog wrote: post #225
OsamaJaber wrote: post #220

The modified MIT clause is sneakier than people think. Hit 100M users or $20M a month and you have to slap "Kimi K2.6" on your UI. That covers any consumer app worth building. Not really open, more like free until you matter. Llama pulled the same move

And the Kimi team broke the Anthropic ToS by training off Opus outputs and… nothing happened?

Nobody cares, nor should they. Anthropic broke nearly every ToS of every website that they scraped data from. The AI robber barons just want to monopolize intellectual property violations, and I'm gonna cheer on any robin hoods that take it back from them.

Re: Kimi K2.6: Advancing open-source coding

#243

Post by greenavocado »

deaux wrote: post #81
greenavocado wrote: post #19

I pray the benchmark figures are true so I can stop paying Anthropic after screwing me over this quarter by dumbing down their models, making usage quotas ridiculously small, and demanding KYC paperwork.

> dumbing down their models, This should be so easy to prove if it were true. Yet there is none of it, just vibes. Still, your other two points are completely valid. The opaqueness of usage quotas is a scam, within a single month for a single model it can differ by more than 2x. And this indeed has been proven.

> This should be so easy to prove if it were true.

https://github.com/anthropics/claude-code/issues/42796

https://scortier.substack.com/p/claude-code-drama-6852-sessi...

Re: Kimi K2.6: Advancing open-source coding

#244

Post by Saline9515 »

OsamaJaber wrote: post #220

The modified MIT clause is sneakier than people think. Hit 100M users or $20M a month and you have to slap "Kimi K2.6" on your UI. That covers any consumer app worth building. Not really open, more like free until you matter. Llama pulled the same move

Attribution is a fair clause in opensource. What is the problem? You are making 20M$ a month thanks to their free work.

Re: Kimi K2.6: Advancing open-source coding

#245

Post by cute_boi »

fg137 wrote: post #162
wolttam wrote: post #145

Earlier quoted context omitted.

Kimi has their own subscription that works basically the same as all the others. https://www.kimi.com/code

At $19/month, hard to see why I want to use Kimi over Claude.

for similar plan i think claude costs like $100 a month?

Re: Kimi K2.6: Advancing open-source coding

#246

Post by BrokenCogs »

FlyingSnake wrote: post #68
simonw wrote: post #53

Accessed via OpenRouter, this one decided to wrap the SVG pelican in HTML with controls for the animation speed: https://gisthost.github.io/?ecaad98efe0f747e27bc0e0ebc669e94... Transcript and HTML here: https://gist.github.com/simonw/ecaad98efe0f747e27bc0e0ebc669...

At this point drawing these Pelicans must be in the training data sets.

Yes we all know that, but we still like to see the pelicans because it's a tradition more or less

Re: Kimi K2.6: Advancing open-source coding

#247

Post by airstrike »

charcircuit wrote: post #185
Strom wrote: post #168

Earlier quoted context omitted.

It's tradition at this point. Based on the upvotes the comment receives, it looks like many readers find value in it.

[flagged]

> Please don't post comments saying that HN is turning into Reddit. It's a semi-noob illusion, as old as the hills.

https://news.ycombinator.com/newsguidelines.html

Re: Kimi K2.6: Advancing open-source coding

#248

Post by throwaw12 »

OsamaJaber wrote: post #220

The modified MIT clause is sneakier than people think. Hit 100M users or $20M a month and you have to slap "Kimi K2.6" on your UI. That covers any consumer app worth building. Not really open, more like free until you matter. Llama pulled the same move

if you reach that numbers, kimi would be your least of worries

Re: Kimi K2.6: Advancing open-source coding

#249

Post by phainopepla2 »

fg137 wrote: post #162
wolttam wrote: post #145

Earlier quoted context omitted.

Kimi has their own subscription that works basically the same as all the others. https://www.kimi.com/code

At $19/month, hard to see why I want to use Kimi over Claude.

Claude usage at $20 is basically unusable for serious work. I haven't used Kimi but I'd have to imagine they're offering a good deal more usage for the same price.

Re: Kimi K2.6: Advancing open-source coding

#250

Post by Flux159 »

ninjahawk1 wrote: post #198

I often wonder if in the future, the same way early computers used to take up an entire room but now fit in your pocket, if in the future the equivalent of a data center will be a single physical device like a phone nowadays. And if that’s the case, would it happen much quicker since technology has been speeding up year by year?

There’s some early work being done here by companies looking at making LLM ASICS like Taalas (HC1 gets 17k t/s for llama 8b - currently at 2.5kW which is closer to a single server, but this is their first chip).

There’s other options like photonic computing which might be able to reduce power significantly but are still in research as far as I can tell. Because so much money is invested in AI & traditional gpu inference is so power hungry, I would expect significant improvements in this space quickly.

Post reply on HN