Open-source AI is booming, massively impactful for progress, competition, transparency & orders of magnitude less dangerous than closed-source frontier AI. Time to support it to the max!

Thank you for fighting for open source. Hero move. •

The part that stuck with me: open source lets people own AI instead of renting it from a few big labs. Regulate it like the frontier and you hand that ownership right back to them.

Thanks Clem! It's crazy to contend that concentrating power, behind one or two monoliths, financially motivated, who gate everything behind APIs, and control access as they please... is the best move for society.

I totally agree. It's time to make access to digital intelligence a human right

What do you think of the new Meta model? I think the 9-person training is interesting so far.

The future is open source

Enterprises can no longer afford to ignore open-source models, which offer three key advantages AI self reliance, lower costs, and unmatched data privacy.

@huggingface Clem walks around with huge nuts.

The CEO of the world's biggest open-source model hub says open-source AI is great. Obvious conflict of interest. Also completely right.

that's true but open-sourcing ai also means you're relying on strangers to correct your model before deployment doesn't always work as plann

@huggingface Awesome stuff but prob time to shave that beard. I think people around you might be being too nice.

the ceo of huggingface saying open source is good is breaking news on bloomberg

@huggingface open source model or open source learning data? neural logics is same everywhere, difference only in the inbound data and weight tuning.

Artificial Good Enough Intelligence has four axes: capability, affordability, availability, and sovereignty. Put simply, most users (including companies that use AI in their products) need models that are capable, affordable, available, and under their own control. Optimizing for those criteria shifts attention from raw benchmark supremacy (although benchmarks are increasingly unreliable anyway) toward cost, accessibility, and ownership. h/t @deseventral

@huggingface 🎯

开源 AI 的价值不只是低成本，更是让更多人能看懂、改动和验证。我做小项目时最喜欢这种可拆开的积木，孩子也能一起看懂每一层在做什么。

Open source supports progress, but rapid capability growth may still outpace societal readiness.

Open-source AI driving progress and transparency is important!

Dobby drove 650K+ NFT mints, chat handled 120K+ requests in a week. when tools stay inspectable people actually use them

The open weights conversation is way ahead of the open data one. We can download a model and barely know what it learned from, or who decided what counted as good data in the first place. Open datasets are the harder fight, and the human judgment baked into them is the part that actually buys you transparency.

The real wins will come from small teams wiring open models into agentic loops that solve specific user problems faster than any monolithic lab.

aggregate all the efforts of OS AI man, that is the best way to compete with this proprietary lab models.

Open source = innovation on steroids. Fully agree.

This just means creating two forks for AI - one for the plebs and one closed for the govt / elites

Let them do whatever they want with frontier models, but leave the open source ecosystem alone, with decent to good models. Many of us just want to use AI to improve our business process, workflows and productivity, not create bombs or virus.

Open-source AI fuels safer, smarter progress.

@huggingface Do some more hiring Clem invest more bring a office to SF

@grok why does China block Claude and ChatGPT? Would you recommend China LLM as a safe alternative to feed your data to?