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

GLM-5.2 is the new leading open weights model on Artificial Analysis

artificialanalysis.ai

Post reply on HN
Print view View original

471–476 of 476 posts

Re: GLM-5.2 is the new leading open weights model on Artificial Analysis

#471

Post by r-w »

carter2099 wrote: post #460
chartpath wrote: post #383

Earlier quoted context omitted.

Are there any indications that this will be possible? Consumer hardware will continue getting better but I can't see 512GB RAM in a MacBook Pro any time soon. I'm hoping linear attention techniques plus MoE will make breakthroughs in size/compression and throughput.

> but I can't see 512GB RAM in a MacBook Pro any time soon Could totally see this being a comment from a forum in like 1994 but swap out GB for MB and MacBook Pro to whatever the popular consumer pc was at the time

Yeah but the price of RAM wasn't increasing at that point.

Re: GLM-5.2 is the new leading open weights model on Artificial Analysis

#472

Post by stymaar »

jgalt212 wrote: post #463
stymaar wrote: post #438

Earlier quoted context omitted.

RAM price don't change anything. You can't fit an infinity of tokens in the KV cache, but the ones that are in there when a user request them are still practically free.

so the utility of the KV cache has nothing to do with available RAM?

This proposition has no logical link with any of the above. Read again.

Re: GLM-5.2 is the new leading open weights model on Artificial Analysis

#473

Post by ttul »

slagfart wrote: post #440
ttul wrote: post #171

Earlier quoted context omitted.

DeepSWE “feels” like the right benchmark in comparison to Artificial Analysis indices and other coding benchmarks. And by their metrics, GPT-5.5 is still king in token efficiency, speed, and overall intelligence per dollar. https://deepswe.datacurve.ai/ Fable 5 is cool and all, but we have not yet seen GPT-5.6.

GLM5.2 isn't even on this benchmark

True. Z.AI ran that bench themselves and report 46.2, which is lower than GPT-5.5 and Opus 4.8, but crushing the other open weights models.

https://z.ai/blog/glm-5.2

Re: GLM-5.2 is the new leading open weights model on Artificial Analysis

#474

Post by rsanek »

ComputerGuru wrote: post #241
orbital-decay wrote: post #176

Earlier quoted context omitted.

They have an "exacto" category with providers they supposedly verified

That’s only for tool use.

That's just one part of it. According to https://openrouter.ai/docs/guides/routing/model-variants/exa...

  We use three classes of signals:
   * Tool-calling success and reliability from real traffic
   * Provider performance metrics such as throughput and latency
   * Benchmark and evaluation data as it becomes available

Re: GLM-5.2 is the new leading open weights model on Artificial Analysis

#475

Post by TranquilMarmot »

deadbabe wrote: post #403
chartpath wrote: post #383

Earlier quoted context omitted.

Are there any indications that this will be possible? Consumer hardware will continue getting better but I can't see 512GB RAM in a MacBook Pro any time soon. I'm hoping linear attention techniques plus MoE will make breakthroughs in size/compression and throughput.

There will be a 1024GB unified memory MacBook Pro.

Not at a price that your average consumer can afford for a long, long time

Re: GLM-5.2 is the new leading open weights model on Artificial Analysis

#476

Post by slig »

kristopolous wrote: post #128
slig wrote: post #115

Earlier quoted context omitted.

Thanks for sharing. I'm curious: why didn't you sort with the score descending?

Because it's currently 511 lines. Why would I want to scroll up to see the stuff I care about? Don't you want the relevant stuff to be right there in front of you?

Thank you, that makes a lot of sense.
Post reply on HN