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

Meta Llama 3

llama.meta.com

Post reply on HN
Print view View original

621–630 of 965 posts

Re: Meta Llama 3

#621

Post by krisoft »

HarHarVeryFunny wrote: post #596
simonw wrote: post #204

Earlier quoted context omitted.

Those additional restrictions mean it's not an open source license by the OSI definition, which matters if you care about words sometimes having unambiguous meanings. I call models like this "openly licensed" but not "open source licensed".

Call it what you will, but it'd be silly if Meta let these 700M+ customer mega-corps (Amazon, Google, etc) just take Meta models and sell access to them without sharing revenue with Meta. You should be happy that Meta find ways to make money from their models, otherwise it's unlikely that they'd be giving you free access (until your startup reaches 700M+ customers, when the free ride ends).

> until your startup reaches 700M+ customers, when the free ride ends

No it doesn’t. The licence terms talk about that those who on the release date of llama3 had 700M+ customers need an extra licence to use it. It doesn’t say that you loose access to it if in the future you gain that many users.

Re: Meta Llama 3

#622

Post by krisoft »

HarHarVeryFunny wrote: post #605
mensetmanusman wrote: post #589

Earlier quoted context omitted.

It's a blob that costs over $10,000,000 in electricity costs to compile. Even if they released everything only the rich could push go.

In today's Dwarkesh interview, Zuckerberg talks about energy becoming a limit for future models before cost or access to hardware does. Apparently current largest datacenters consume about 100MW, but Zuck is considering future ones consuming 1GW which is the output of typical nuclear reactor! So, yeah, unless you own your own world-class datacenter, complete with the nuclear reactor necessary to power the training ru…

On a sufficiently large time scale the real limit on everything is energy. “Cost” and “access to hardware” are mere proxies for energy available to you. This is the idea behind the Kardashev scale.

Re: Meta Llama 3

#623

Post by ChristophGeske »

namanski wrote: post #220
a2128 wrote: post #25

Just got uploaded to HuggingFace: https://huggingface.co/meta-llama/Meta-Llama-3-8B https://huggingface.co/meta-llama/Meta-Llama-3-70B

I just hosted both models here: https://chat.tune.app/ Playground: https://studio.tune.app/

Thanks for the link I just tested them and they also weark in europe without the need to start a VPN. What specs are needed to run these models. I mean the llama 70B and the Wizard 8Bx22 model. On your site they run very nicely and the answears they provide are really good they booth passed my small test and I would love to run one of them locally. So far I only ran 8B models on my 16GB RAM pc using LM Studio but having such good models run locally would be awesome. I would upgrade my ram for that. My pc has an 3080 laptop GPU and I can increase the RAM to 64GB. As I understood it a 70B model needs around 64 GB but maybe only if it quantized. Can you confirm that? Can I run Llama 3 as well as you when I simply upgrade my RAM sticks. Or are you running it on a cloud and you can't say much about the requirements for windows pc users? Or do you have hardware usage data for all the models on your site and you can tell us what they need to run?

Re: Meta Llama 3

#624

Post by SushiHippie »

chown wrote: post #553
gpm wrote: post #189

Earlier quoted context omitted.

From the faq > Does Msty support GPUs? > Yes on MacOS. On Windows* only Nvidia GPU cards are supported; AMD GPUs will be supported soon. Do you support GPUs on linux? Your downloads with windows are also annotated with CPU/CPU + GPU, but your linux ones aren't. Does that imply they are CPU only?

Yes, if CUDA drivers are installed it should pick it up.

> AMD GPUs will be supported soon.

Will AMD support also land on linux?

Re: Meta Llama 3

#625

Post by nalekberov »

jimbokun wrote: post #511
m12k wrote: post #504

Earlier quoted context omitted.

But also: Facebook/Meta got burned when they missed the train on owning a mobile platform, instead having to live in their competitors' houses and being vulnerable to de-platforming on mobile. So they've invested massively in trying to make VR the next big thing to get out from that precarious position, or maybe even to get to own the next big platform after mobile (so far with little to actually show for it at a str…

Similarly to Google keeping Android open source, so that Apple wouldn’t completely control the phone market.

In fact Google doesn't care much if Apple controls the entire mobile phone market, Android is just guaranteed way of acquiring new users. Now they are paying yearly around $19 billion Apple to be default search engine, I expect without Android this price would be times more.

Re: Meta Llama 3

#626

Post by HarHarVeryFunny »

krisoft wrote: post #621
HarHarVeryFunny wrote: post #596

Earlier quoted context omitted.

Call it what you will, but it'd be silly if Meta let these 700M+ customer mega-corps (Amazon, Google, etc) just take Meta models and sell access to them without sharing revenue with Meta. You should be happy that Meta find ways to make money from their models, otherwise it's unlikely that they'd be giving you free access (until your startup reaches 700M+ customers, when the free ride ends).

> until your startup reaches 700M+ customers, when the free ride ends No it doesn’t. The licence terms talk about that those who on the release date of llama3 had 700M+ customers need an extra licence to use it. It doesn’t say that you loose access to it if in the future you gain that many users.

You don't lose access, but the free ride ends. It seems that new licence will include payment terms. Zuckerberg discusses this on the Dwarkesh interview.

Re: Meta Llama 3

#627

Post by ks2048 »

matsemann wrote: post #75
bbig wrote: post #4

They've got a console for it as well, https://www.meta.ai/ And announcing a lot of integration across the Meta product suite, https://about.fb.com/news/2024/04/meta-ai-assistant-built-wi... Neglected to include comparisons against GPT-4-Turbo or Claude Opus, so I guess it's far from being a frontier model. We'll see how it fares in the LLM Arena.

> Meta AI isn't available yet in your country Where is it available? I got this in Norway.

Everyone saying it's an EU problem. Same message in Guatemala.

Re: Meta Llama 3

#629

Post by synergy20 »

tried to run and it needs lots of memory from the low end GPU, would be nice if it has a requirement checklist, the 8B model is about 16GB to download.

Re: Meta Llama 3

#630

Post by alsodumb »

Workaccount2 wrote: post #493
lordswork wrote: post #384

Earlier quoted context omitted.

TikTok (ByteDance) is now building an AGI team to train and advance LLMs (towards AGI), probably after realizing they are in a similar scenario.

I don't know how they think they are going to get the required number of GPU's through export controls.

Are the export controls to China geographically or any Chinese majority-owned entity? Either way, ByteDance has tons of offices everywhere in the world including Singapore, US, etc. Given the money, I don't think GPU access wouldn't be their biggest problem.
Post reply on HN