Not all R1 experiences are equal, here is a ranking of the top few. Tencent Yuanbao really stands out here.
Artificial Intelligence thread
- Thread starter 9dashline
- Start date
"Chinese AI team wins global award for replacing Nvidia GPU with industrial chip
Win for video-generation AI model based on off-the shelf chip marks a first for China and a shift in global race to optimise AI hardware.
In a bold challenge to US giant Nvidia’s dominance in artificial intelligence (AI) hardware, Chinese researchers have trained a cutting-edge video-generation model on an off-the shelf industrial chip – outperforming high-end GPUs in both speed and efficiency.[...]"
Win for video-generation AI model based on off-the shelf chip marks a first for China and a shift in global race to optimise AI hardware.
In a bold challenge to US giant Nvidia’s dominance in artificial intelligence (AI) hardware, Chinese researchers have trained a cutting-edge video-generation model on an off-the shelf industrial chip – outperforming high-end GPUs in both speed and efficiency.[...]"
One of the best visual and technical explanations of how DeepSeek changed the AI landscape for the better
Google was caught lying about the number of GPUs needed to run DeepSeek R1 to portray their own Gemma 3 model more favorably. Like I said before Western AI labs are getting desperate, hence the increasing number of hyperbolic or even false claims.
OAI getting desperate:
The proposal:
Some highlights:
” As America’s world-leading AI sector approaches artificial general intelligence (AGI), with a Chinese Communist Party (CCP) determined to overtake us by 2030, the Trump Administration’s new AI Action Plan can ensure that American-led AI built on democratic principles
continues to prevail over CCP-built autocratic, authoritarian AI.
...
In advancing democratic AI, America is competing with a CCP determined to become the global leader by 2030. That’s why the recent release of DeepSeek’s R1 model is so noteworthy—not because of its capabilities (R1’s reasoning capabilities, albeit impressive, are at best on par with several US models), but as a gauge of the state of this competition. As with Huawei, there is significant risk in building on top of DeepSeek models in critical infrastructure and other high-risk use cases given the potential that DeepSeek could be compelled by the CCP to manipulate its models to cause harm. And because DeepSeek is simultaneously state-subsidized, state-controlled, and freely available, the cost to its users is their privacy and security, as DeepSeek faces requirements under Chinese law to comply with demands for user data and uses it to train more capable systems for the CCP’s use. Their models also more willingly generate how-to’s for illicit and harmful activities such as identity fraud and intellectual property theft, a reflection of how the CCP views violations of American IP rights as a feature, not a flaw."
The proposal:
Some highlights:
” As America’s world-leading AI sector approaches artificial general intelligence (AGI), with a Chinese Communist Party (CCP) determined to overtake us by 2030, the Trump Administration’s new AI Action Plan can ensure that American-led AI built on democratic principles
continues to prevail over CCP-built autocratic, authoritarian AI.
...
In advancing democratic AI, America is competing with a CCP determined to become the global leader by 2030. That’s why the recent release of DeepSeek’s R1 model is so noteworthy—not because of its capabilities (R1’s reasoning capabilities, albeit impressive, are at best on par with several US models), but as a gauge of the state of this competition. As with Huawei, there is significant risk in building on top of DeepSeek models in critical infrastructure and other high-risk use cases given the potential that DeepSeek could be compelled by the CCP to manipulate its models to cause harm. And because DeepSeek is simultaneously state-subsidized, state-controlled, and freely available, the cost to its users is their privacy and security, as DeepSeek faces requirements under Chinese law to comply with demands for user data and uses it to train more capable systems for the CCP’s use. Their models also more willingly generate how-to’s for illicit and harmful activities such as identity fraud and intellectual property theft, a reflection of how the CCP views violations of American IP rights as a feature, not a flaw."
Translation:
"How we can justify our outrageous pricing scams and hot air valuation if Chinese companies are offering everything that we have cheaper and better"
Those evil SeeSeePee dont have respect for IP creators.
Unlike OpenAI that scouted the entire internet, pirated and used every single book, thread, comment, art, videos, paper, article without the author's permission or consent and on top of that use underpaid labor in africa to curate the data to train their stupid models.
Unlike OpenAI that scouted the entire internet, pirated and used every single book, thread, comment, art, videos, paper, article without the author's permission or consent and on top of that use underpaid labor in africa to curate the data to train their stupid models.
OAI's founding principles have longed been abandoned, its top researchers have jumped ship, .
We'll find out in literally a year tops if they still retain their fabled first mover advantage.
In terms of policy on AI and semis, expect deregulation and tighter export controls/student/work visas. Same old, really.
Here's another banger:
That's one way to admit GPT4.5 is actually GPT5