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

Hugging Face raises $235M from investors including Salesforce and Nvidia

techcrunch.com

Post reply on HN
Print view View original

101–110 of 207 posts

Re: Hugging Face raises $235M from investors including Salesforce and Nvidia

#101

Post by brucethemoose2 »

steno132 wrote: post #61

Congratulations! Elon's team over at X.ai, who I'm helping advise, is choosing between Hugging Face and a locally hosted alternative. What would HN recommend? I prefer Hugging Face as it has a stronger community built in but others prefer a open source project we can customize.

Why not both?

Y'all could host models on HF, and use the HF format/download code (which is quite good).

But make a pretty frontend for it.

Down the road you could just host your own backend implementing the HF API, if such a thing is necessary.

Re: Hugging Face raises $235M from investors including Salesforce and Nvidia

#102

Post by version_five »

calebkaiser wrote: post #64
version_five wrote: post #45

Earlier quoted context omitted.

I agree and so far I think it's picks and shovels all the way down. Most of the "AI" companies I see sell tooling, often to "help your AI team work faster|better|spend less time doing x|whatever. It all presupposes there are end uses, which are few and far between, certainly almost none are being sold. AI is still in the hype mode of companies having internal budget to "invest in AI" and so that's who the AI companie…

It's important to note that despite the recent excitement about LLMs, which is still an emerging technology, "AI" is not a new market by any means, nor are major companies only now investing in it. For the better part of a decade, ML has been widely adopted across industries, and the average person uses an "AI" system many times in a given day. For example, if you open the home screen on the average smartphone right…

That's the framing we've been using for years, look at some big tech ML or data science that we reframed as AI, your manufacturer/transportation/retail/whatever company needs it too, but there are no products for end uses in your industry, just lots of tooling and consulting services you can buy for bespoke projects.

Re: Hugging Face raises $235M from investors including Salesforce and Nvidia

#103

Post by emmender1 »

alberth wrote: post #65
emmender1 wrote: post #44

Earlier quoted context omitted.

you are: a) a hardware vendor nvidia/amd whose products are needed by anyone in the game b) you have a captive customer base already ( microsoft, salesforce, servicenow, adobe) to whom you can sell ai/ml value adds c) you make money via ads (google facebook) and ai/ml helps with better targeting everyone else is pissing away VC money.

A) so you’re competing against aws, azure, etc. That sounds brutally tough. B) re: value adds, like what exactly? C) this seems like such core functionality that a company wouldn’t outsource this to a 3rd party vendor. If that’s the case, there isn’t an opportunity to sell anything if you’re that AL/ML vendor then.

A) aws/azure need nvidia/amd chips in their servers... it is too difficult to homegrow these chips (altho aws is trying) B) microsoft (github autopilot 20$ per month per user), document/image autogeneration from a prompt (likely in word doc 5$ a month per user for this) C) right yea, thats why fb/goog have big ai farms

Re: Hugging Face raises $235M from investors including Salesforce and Nvidia

#104

Post by version_five »

brucethemoose2 wrote: post #95
jeroenhd wrote: post #51

Earlier quoted context omitted.

Their Git LFS based hosting is used by a lot of AI tools. Huggingface level data storage and transfer would also be VERY costly to accomplish for a small party. Some of these models are several gigabytes in size and downloaded hundreds or thousands of time per day. What they do isn't too hard to replicate, but the price for which they do it is impossible to compete with.

CivitAI has somehow managed this over in Stable Diffusion land. In fact, I think the whole community is HF averse. The two most popular frameworks are based around the Stability implemenetaion and file format, not HF diffusers.

Any sense of why they are HF averse?

Re: Hugging Face raises $235M from investors including Salesforce and Nvidia

#105

Post by asadm »

alberth wrote: post #68
asadm wrote: post #60

Earlier quoted context omitted.

They will be acquired. Either by Google or MS.

MS has already invested $10B in OpenAI, so they aren’t going to be a buyer. Google has poured hundreds of million into Anthropic due to Eric Schmidt, so they aren’t going to be a buyer.

Both Anthropic and OpenAI are not a replacement for huggingface. Which has complete monopoly on model hosting right now.

Re: Hugging Face raises $235M from investors including Salesforce and Nvidia

#106

Post by zo1 »

Lord_Zero wrote: post #76
michaelt wrote: post #73

Earlier quoted context omitted.

Pip, maven, npm, docker hub, apt-get, github and suchlike seem to manage to provide hefty binary downloads with no need to monetise anonymous downloads. I don't quite understand how they do so, but it seems to be possible. Everyone who downloads a 5GB stable diffusion model probably downloaded 5GB of pytorch+cuda+cudnn first.

Back in the day, universities would host all that shit via their mirrors for free for everyone. Not sure where it all is today with the cloud being so prevalent.

You don't even hear the term mirror anymore.

Re: Hugging Face raises $235M from investors including Salesforce and Nvidia

#107

Post by dbish »

ladberg wrote: post #41
taminka wrote: post #33

Earlier quoted context omitted.

i'm curious what HF does, because iiuc they're mostly just hosting model files? i know that they have some compute offers, and also created/maintain a few libraries, but it's not particularly widely used, i'm really not sure how they're supposed to earn money...

I think their "spaces" (i.e. running tons of models for free with a nice UI) are widely used.

This isn’t sustainable though. Compute is expensive, maybe they’ll end up with paid plans for anyone trying to use a space

Re: Hugging Face raises $235M from investors including Salesforce and Nvidia

#108

Post by dbish »

fidotron wrote: post #12

ML models are the new apps. There is a huge opening for an App Store type situation which enables people to buy models and integrate them into their products, handling the appropriate licensing. Bonus points for certifying the models actually do what they say. That by itself will probably become a mini industry. Hugging Face are the clear leaders in terms of having mindshare in the community to be able to build it.

Models != apps, they’re closer to the backend for apps or a core library, so your customers aren’t “everyone with a phone/computer” like an app, it’s the much smaller, but potentially high impact, “everyone who creates apps”.

We’re not in a world where non tech people are searching for models in the model store from their phone.

Re: Hugging Face raises $235M from investors including Salesforce and Nvidia

#109

Post by hereonout2 »

taminka wrote: post #33
QuadrupleA wrote: post #16

Hugging Face is early in the Silicon Valley enshittification cycle - currently burning VC money being incredibly good to it's users. Next would be shifting that value to its business customers. Then clawing back and squeezing as much value as possible from those. Then collapse. Although, is it technically a platform / marketplace? Anyway for now, enjoy the free beer from the VCs.

i'm curious what HF does, because iiuc they're mostly just hosting model files? i know that they have some compute offers, and also created/maintain a few libraries, but it's not particularly widely used, i'm really not sure how they're supposed to earn money...

I've been asking myself that for a while. Would they be so popular if they didn't host & serve petabytes of model files for free? How will they monetize that aspect to match this valuation?

Whilst the scale of their model hosting is impressive, the functionality seems pretty basic. The models are just BLOBs in git LFS repos, you're usually relying on knowing which users to follow, then learning the way they name their models and how that naming convention applies to the particular framework and hardware you're using.

As an example the user "TheBloke" is prolific at publishing LLM models for various hardware / framework combos, but look how little the HF interface actually helps navigate or find what you're after: https://huggingface.co/TheBloke

Also, is anybody using HF Hub in "production"? We've deployed a few LLMs now and once we've decided on a model the first thing we do is get it off HF and into our own storage ready for deployment. There seems no reason to tightly integrate HF Hub with production systems given it's a just a bunch of files you can copy and keep.

Re: Hugging Face raises $235M from investors including Salesforce and Nvidia

#110

Post by dbish »

steno132 wrote: post #61

Congratulations! Elon's team over at X.ai, who I'm helping advise, is choosing between Hugging Face and a locally hosted alternative. What would HN recommend? I prefer Hugging Face as it has a stronger community built in but others prefer a open source project we can customize.

If you’re hosting for your own use, unclear why you wouldn’t just do that directly on cloud instances. If you’re trying to share models with others publicly, then HF is the best option right now.
Post reply on HN