Why NVIDIA Acquired Hugging Face for $12.9B | Thomas Wolf

Thomas Wolf (0:00)

I think we're just amazingly aligned. Everyone thinks it's great news for the open-weights ecosystem. And I think the whole community at Hugging Face has been very excited about that. Right now, we have about 17 million users.

We passed about 3 million models. Our goal is to keep growing the open-weights ecosystem. Working with Nvidia, who is amazingly aligned with us, is the best way to do that.

SPEAKER_2 (0:30)

And we are back. We are live with Thomas Wolf, who is the co-founder and chief science officer of Hugging Face, which was recently acquired by Nvidia for 12.9... Was it $12.9303 billion, which is amazing. Thomas, congratulations and welcome to MTS.

Thomas Wolf (0:50)

Yeah.

SPEAKER_2 (0:50)

Thanks. Happy to be here. So tell us more about the Nvidia acquisition.

Why did Hugging Face decide to join Nvidia?

Thomas Wolf (0:58)

Yeah. I mean, I guess you've been following me like me, what Jensen is speaking about, the open source, the open weights model. Later, that was his first tweet ever that became viral. He's pretty good at getting every tweet viral since then. But I think we're just amazingly aligned. And I think the reaction to when we announced, the acquisition, which is now feel like ages ago in AI lifetime, but like was just two or three weeks ago now. Just say it all, I think everyone was very, very excited.

Everyone I think around from Satya, Sandar, just to name some of the main person who responded to our announcements. Everyone thinks it's great news for the open weights ecosystem. And I think, you know, general, the whole community at Hugging Face as well has been very excited about that. There's a notion that we can do a lot more. So right now we have this about 17 million users. You know, we passed about 3 million models. Our goal is to keep growing the open weights ecosystem. We think we are not even halfway there, I would say. I feel like open source could be way larger in terms of being able to defuse AI in the community. And we think working with Nvidia, who is amazingly aligned with us, is the best way to do that.

We also share many things, I mean, in terms of robotics, in terms of giving the tools, the community to companies to train our models, you know, NemoTron, you know, we also have our own effort at Hugging Face to give the tools, to give the pre-training datasets, to give teaching around how to train models. Everything here is so well aligned. I think it's going to be a great win for the open source community and the open weights community. And I hope, I mean, my hope is still that this open weights committee is going to be extremely large. We think the future of AI, hopefully will be more than just two companies selling tokens to everyone. It will be more a lot of companies training their own models, a lot of companies being able to own their own to the best. And that usually means using open weights model.

SPEAKER_3 (3:20)

Yeah, totally. What do you think will be the pivotal point where more people start actually using open weight models or more open source technology?

Thomas Wolf (3:34)

I might be slightly biased, but I feel like we're starting to see the first hint this year and this summer.

I feel like it started probably in the beginning of the year with a couple of companies. I remember Uber and Meta saying that the token budget they were spending were too high, and so they were starting to investigate open weights model. I think for a long time, the main challenge here was that fine tuning an open weight model for your internal use was a bit of a dark science, dark magic, alchemy.

But this is changing for two reasons. The first reason is you start to have a lot of teams that are doing that for us. Our friends at Prime Intellect, their main job is doing that. Trajectories, they're basing. A lot of teams are starting to help companies doing that. The other thing is, even if you don't want to pay another team, you can start to ask CloudCode, or you can start to ask ChatGPT to actually help you. So in one way, this is the beginning of RSI, which is an agent and a model helping you to train a model. So it can be to improve this model, but it can also be helping you to train an open-weight model for your specific internal task. So you start to see some indicators. So if you follow, for instance, Vercel, they could list how much tokens are used in the API between the open-weight model, open-weight API, inference API, and the closed-weights.