The founder of Black Myth Wukong commented on DeepSeek: It may be a national technological achievement
DeepSeek has recently sparked discussion, and the founder of Black Myth Wukong has evaluated DeepSeek: it may be a technological achievement at the level of national destiny. The following is the full text: Let's seriously promote DeepSeek before the new year. V3 has been used for a month, but R1 has only been used for 5 days, and I have to post this Weibo. First of all, a violent argument: DeepSeek may be a technological achievement at the level of national destiny. In order to make it clear how amazing this achievement is, let me give an example: if an AI model can achieve any of the following, it is a super breakthrough - 1. Powerful. Comparable to O1's reasoning ability, there is currently no other model that can match it. 2. Cheap. Fewer parameters, training costs and usage fees are reduced by an order of magnitude. 3. Open source. Anyone can download and deploy it by themselves, providing detailed instructions on how to train and tips in papers, and even providing mini models that can run on mobile phones. 4. Free. The services provided by the official are completely free, and anyone can use them anytime, anywhere. 5. Networked. The only inference model that currently supports networked search (O1 does not yet support it). 6. Local. DeepSeek is a young Chinese company with a small scale, developed by a local team without overseas experience or senior industry experience. DeepSeek has achieved all of the above six points at the same time. It's too amazing, it's unreal. If this is not worth cheering, what else is worth cheering for? So, what should you do when you see this Weibo now? 1. Please visit directly and start using it immediately, and there is also a mobile APP. 2. Use it hard, use it crazy, and try to use it to replace traditional search. Imagine that you already know the top experts in every field, and they are all your good friends who are online at any time. Take all the questions that may need to "consult others" and ask it first - whether it is to build a complex function, figure out which swimming pool nearby is the best, check the legal pitfalls in the contract, make the newly written lyrics rhyme better, or help parents confirm whether a certain health product is really effective. 3. Go and see how others use it, try other large models, understand what AI is good at, what it is not good at, how to tune it, and then continue to unlock and iterate your own usage and more tools. I hope DeepSeek R1 will make you demystify the most advanced AI currently, and gradually turn AI into water and electricity in your life. So lucky! So happy! Such a shocking breakthrough comes from a purely Chinese company.
Disclaimer: The content of this article solely reflects the author's opinion and does not represent the platform in any capacity. This article is not intended to serve as a reference for making investment decisions.
You may also like
Arizona Senate Approves Bitcoin Reserve Bill: A Leap Towards Crypto Dominance in the U.S?
Trailblazing or Isolated Incident? Evaluating Arizona's Groundbreaking Bitcoin Reserve Bill and Its Potential Ripple Effect Across U.S. States
DeepSeek’s Debut Stirs Crypto Market, BTC ETFs Drop by $457M – Insights Revealed
DeepSeek's Entry Culminates in $534M Crypto Outflows, Shaking BTC and Ethereum ETFs While XRP Shows Resilience
Ethereum’s Active Users Soar 37%: Are Institutions and DeFi Driving Growth?
Unpacking the Factors Behind Ethereum's Rise: Examining the Role of DeFi and Institutional Investment