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

Kimi K2.5 Technical Report [pdf]

github.com

Post reply on HN
Print view View original

71–80 of 146 posts

Re: Kimi K2.5 Technical Report [pdf]

#71

Post by jychang »

zzleeper wrote: post #70

Do any of these models do well with information retrieval and reasoning from text? I'm reading newspaper articles through a MoE of gemini3flash and gpt5mini, and what made it hard to use open models (at the time) was a lack of support for pydantic.

That roughly correlates with tool calling capabilities. Kimi K2.5 is a lot better than previous open source models in that regard.

You should try out K2.5 for your use case, it might actually succeed where previous generation open source models failed.

Re: Kimi K2.5 Technical Report [pdf]

#73

Post by m3kw9 »

epolanski wrote: post #15

It's interesting to note that a model that can OpenAI is valued almost 400 times more than moonshotai, despite their models being surprisingly close.

Unless they can beat their capabilities by a clear magical step up and has infrastructure to capture the users

Re: Kimi K2.5 Technical Report [pdf]

#74

Post by Tepix »

gedy wrote: post #59

Sorry if this is an easy-answerable question - but by open we can download this and use totally offline if now or in the future if we have hardware capable? Seems like a great thing to archive if the world falls apart (said half-jokingly)

You could buy five Strix Halo systems at $2000 each, network them and run it.

Rough estimage: 12.5:2.2 so you should get around 5.5 tokens/s.

Re: Kimi K2.5 Technical Report [pdf]

#75

Post by nl »

Imustaskforhelp wrote: post #28
eknkc wrote: post #11

Earlier quoted context omitted.

I've been using it with opencode. You can either use your kimi code subscription (flat fee), moonshot.ai api key (per token) or openrouter to access it. OpenCode works beautifully with the model. Edit: as a side note, I only installed opencode to try this model and I gotta say it is pretty good. Did not think it'd be as good as claude code but its just fine. Been using it with codex too.

I tried to use opencode for kimi k2.5 too but recently they changed their pricing from 200 tool requests/5 hour to token based pricing. I can only speak from the tool request based but for some reason anecdotally opencode took like 10 requests in like 3-4 minutes where Kimi cli took 2-3 So I personally like/stick with the kimi cli for kimi coding. I haven't tested it out again with OpenAI with teh new token based pri…

I like Kimi-cli but it does leak memory.

I was using it for multi-hour tasks scripted via an self-written orchestrator on a small VM and ended up switching away from it because it would run slower and slower over time.

Re: Kimi K2.5 Technical Report [pdf]

#76

Post by syndacks »

How do people evaluate creative writing and emotional intelligence in LLMs? Most benchmarks seem to focus on reasoning or correctness, which feels orthogonal. I’ve been playing with Kimmy K 2.5 and it feels much stronger on voice and emotional grounding, but I don’t know how to measure that beyond human judgment.

Re: Kimi K2.5 Technical Report [pdf]

#77

Post by j-bos »

Tepix wrote: post #74
gedy wrote: post #59

Sorry if this is an easy-answerable question - but by open we can download this and use totally offline if now or in the future if we have hardware capable? Seems like a great thing to archive if the world falls apart (said half-jokingly)

You could buy five Strix Halo systems at $2000 each, network them and run it. Rough estimage: 12.5:2.2 so you should get around 5.5 tokens/s.

Is the software/drivers for networking LLMs on Strix Halo there yet? I was under the impression a few weeks ago that it's veeeery early stages and terribly slow.

Re: Kimi K2.5 Technical Report [pdf]

#79

Post by fragmede »

gedy wrote: post #59

Sorry if this is an easy-answerable question - but by open we can download this and use totally offline if now or in the future if we have hardware capable? Seems like a great thing to archive if the world falls apart (said half-jokingly)

Yes but the hardware to run it decently gonna cost you north of $100k, so hopefully you and your bunkermates allocated the right amount to this instead of guns or ammo.
Post reply on HN