Grok3 Launch [video] • Hacker News Board

Re: Grok3 Launch [video]

#491

Post by phillipcarter » Tue, Feb 18, 2025, 1:58 PM UTC

3abiton wrote: post #74

1024core wrote: post #39

Earlier quoted context omitted.

And Anthropic not even in the top 10 ...

I keep hearing about Claude's impressive coding skills (compared to its benches) yet, not evident for me (I use the web version, not cline). Compared to 4o it's not that great.

What are you using it for in general? IME the reason Claude pulls out ahead is that when you use it in a larger existing codebase, it keeps everything "in the style" of that codebase and doesn't veer off into weird territory like all the others.

Re: Grok3 Launch [video]

#492

Post by Cheer2171 » Tue, Feb 18, 2025, 1:59 PM UTC

ban-evader wrote: post #436

croes wrote: post #170

Earlier quoted context omitted.

Who cares about benchmarks? These things still cost me time because of hallucinations.

You’re a very poor user of LLMs if they’re not a net time saver for you.

So the No True Scotsman fallacy of LLM productivity?

Re: Grok3 Launch [video]

#493

Post by bearjaws » Tue, Feb 18, 2025, 2:00 PM UTC

pveierland wrote: post #200

TLDW. Will this be open weights? This commit seems to indicate so, but neither HF or GH has public data yet: https://huggingface.co/xai-org/grok-1/commit/91d3a51143e7fc2... Edit: Answer from Elon in video is that they plan to make Grok 2 weights open once Grok 3 is stable.

This is how they've done the past releases as well, soon after they release the latest and greatest they open source the last model.

Re: Grok3 Launch [video]

#494

Post by abrahamepton » Tue, Feb 18, 2025, 2:00 PM UTC

[flagged]

Re: Grok3 Launch [video]

#495

Post by bearjaws » Tue, Feb 18, 2025, 2:01 PM UTC

1024core wrote: post #39

Rover222 wrote: post #13

Grok 3 at the top of Chatboat Arena with 1400, and the model will continue to improve as it trains more.

And Anthropic not even in the top 10 ...

Any model that censors itself does poorly, despite being able to provide high quality answers.

Re: Grok3 Launch [video]

#496

Post by abrahamepton » Tue, Feb 18, 2025, 2:02 PM UTC

yuppii wrote: post #469

makerofthings wrote: post #139

[flagged]

[flagged]

You can’t name one single lie. If so, please do.

Re: Grok3 Launch [video]

#497

Post by mrkstu » Tue, Feb 18, 2025, 2:03 PM UTC

gordon_freeman wrote: post #235

namaria wrote: post #179

Earlier quoted context omitted.

The only plausible explanation for the amount of resources poured into these language models is the hope that they somehow become the origin of AGI, which I think is pretty fanciful. I can feel the cold wind of the next AI winter coming on. It's inevitable. Computers are good at emulating intelligent behavior, people get excited that it's around the corner, and the hype boils over. This isn't the last time this will…

Everyone seems to have a different definition for AGI. Is there some kind of standard there?

No- but the main issue is that all reasonable ones I can conceive lead inevitably to the Singularity technologically, and pretty quickly since we seem determined to throw as much silicon as possible at the problem. Hopefully the final step is intractable.

Re: Grok3 Launch [video]

#498

Post by KTibow » Tue, Feb 18, 2025, 2:04 PM UTC

lionkor wrote: post #144

I don't understand how and why Grok would be related to "understanding the nature of the universe", as Musk puts it. Please correct me if I'm wrong, but they basically just burned more cash than any human should have to buy Nvidia GPUs and make them predict natural language, right? So, they are somewhat on-par with all the other companies that did the same. This is not innovation, this is baseless hype over a mediocr…

It's not much better than DeepSeek's old slogan "Unravel the mystery of AGI with curiosity. Answer the essential question with long-termism."

Re: Grok3 Launch [video]

#499

Post by jamalaramala » Tue, Feb 18, 2025, 2:05 PM UTC

golol wrote: post #442

yodsanklai wrote: post #421

Earlier quoted context omitted.

Naive question from a bystander , but since DeepSeek is open source and is on par with o1-pro (is it?), shouldn't we expect that anybody with the computer power is capable to compete with o1-pro?

Deepseek is not on par with o1.

It probably depends on the benchmark you choose; according to Chatbot Arena, Deepseek-R1 ranks similarly to o1-2024-12-17; and Grok3 is just 3% above these models in "Arena Score" points.

Re: Grok3 Launch [video]

#500

Post by itishappy » Tue, Feb 18, 2025, 2:06 PM UTC

vachina wrote: post #419

greatgib wrote: post #310

Billions spent, one of the most powerful AI developed, and still no one competent enough to trim the 15 mins of waiting time filler at the beginning of the announcement video...

Tells me they have spent their entire engineering time on engineering and zero on marketing fluff, which is good.

Not sure I share the same takeaway from their marketing video.