“Nvidia offered an investment at a $7 billion valuation, which is different from an acquisition. I don't know if we know how much money Nvidea was offering, or how big of a share of Hugging Face they would have owned after the investment.”
“Hardware + CUDA + inference + licensing + potentially Hugging Face. You seriously don’t see the monopoly risk here? That’s way too much control in one company.”
“If hugging face is primarily focused on selling monthly accounts rather than token based billing which bills more as the company’s expenses increase, its not likely to ever be sustainable without significant changes.”
“If I am right, HF is dead in the water. 2026 is the last of the free to download models. Instead of almost weekly flooding of model news, it will become a drip in an extreme case.”
“The business plan seems like it's not got a huge moat - or is there something I'm missing? So is this play just because competing with hugging face is bad PR to the tune of this price tag; or because they're afraid it'll get acquired by somebody else (AMD?”
“Hugging Face was the ultimate neutral marketplace for AI. Now, that marketplace belongs to AMD’s fiercest competitor. Nvidia can subtly dictate which hardware gets first-class support.The CUDA Lock-In Deepens.”
“I have the same, I highly recommend using the MTP Qwen3.6 35B A3B from unsloth at Q5 (https://huggingface.co/unsloth/Qwen3.6-35B-A3B-MTP-GGUF). I run using llama.cpp.”
“The hard half is that `from_pretrained("org/model")` resolves to huggingface.co by default, and that string is sitting in thousands of repos, Dockerfiles, notebooks and papers. Forking transformers takes an afternoon.”
“> Hugging Face has essentially become github for model repositories and I think nvidia wanted to snag that before microsoft does.Given that most of the weights are from Chinese labs and teams, I'm surprised China's own ModelScope hasn't leapfrogged HuggingFace.”
“For anyone running infrastructure on top of this: does a change of ownership at Hugging Face actually move the needle on model availability, pricing, or hosting terms? Or does the neutral-hub reputation just get harder to keep once one shareholder has an obvious stake in the outcome?”
“Grim prediction:"Quantized models are no longer permitted on the Hugging Face platform. Reduced precision models are a safety hazard and a violation of our TOS. Click here to speak with a sales representative about our many exciting cloud hosting offerings or enterprise GPU packages."”
“If Nvidia owns Hugging Face, they have an obvious incentive to make the ecosystem run exceptionally well on Nvidia hardware. That could show up in optimization priorities, tooling, benchmarks, deployment recommendations, or which models get the most attention not necessarily by blocking alternatives.”
“\~You have to appreciate an AI company that calls itself Hugging Face. It sounds warm and friendly, but then you remember what a face hugger actually is...”
“To play devil's advocate: Hugging Face was probably not profitable, and someone has to pay the bills long-term. Nvidia benefits from Hugging Face and the operating costs are pennies to them.”
“hugging face is basically a real easy way to run llms. They also provide a bunch of libraries to do things like split compute across all your resources.”
“The only thing I can see them being able to get away with is increasingly bending the hugging face python API and any other features of that sort they develop to NVIDIA only.”