Mstar retweeted
короче, вот вам самый полный обзор на два дня саммита России, США и Китая:
- Путин похвалил Си и Трампа
- Си похвалил Путина и Трампа
- Трамп похвалил Путина, Си и Трампа
感谢华为
DeepSeek-V4.1-Flash was getting surprisingly low benchmark scores on Huawei's new Ascend A5 chips, and now we know why.
A bug in vLLM-Ascend was causing decoder layers 21–39 to use the wrong MoE expert weights, basically reusing weights from layers 0–18.
After fixing a single layer counter, its GSM8K score jumped from 75.22% to 97.38% in just one test.
That's a 22.16 percentage-point improvement from fixing one bug, not changing the model.
PR: github.com/vllm-project/vllm…
Mstar retweeted
Over the past week, I’ve been overwhelmed by the love and support of friends, industry colleagues, and people I've never met. Thank you to everyone who has shown kindness to my family and me. It means the world to me.
I came to the United States 53 years ago, and becoming an American citizen is one of the greatest gifts and proudest moments of my life. NVIDIA has been my life’s work. Building it into the company it is today alongside my colleagues has been a privilege and an honor. Nowhere but in America could a company like NVIDIA be possible. I love our country, and I love NVIDIA.
I did not live up to the standards expected of everyone at NVIDIA, and I have made serious mistakes in my personal life. I created problems for the Board that I deeply regret. I own my mistakes completely, and I am committed to cooperating with the Board’s independent investigation.
Since my departure, I’ve seen an outpouring of calls from NVIDIA’s employees, investors, business partners, and customers for me to return as CEO. Many continue to reach out to me directly as the company navigates complex technology, market, competitive, and geopolitical dynamics. NVIDIA is a once-in-a-generation company. If the Board of Directors decides that my return is in the best interest of our shareholders, our employees, the company, and America, I would be honored to answer these calls and return as CEO.
Thank you again to everyone who has shared their good wishes. I am profoundly humbled by your support. To my family: I love you, and I will spend the rest of my life being worthy of yours.
Jensen
谁能不支持打百年战争呢?
“不在餐桌上就在菜单上”是老生常谈,但在之前的日本没几个人知道这句话。这几天总算有了点热度。
这句话用来描写雅尔塔会议最合适。桌上的英美苏不但决定了如何分割德日战败国,就连同属战胜国的中国因为没在餐桌上就也在了菜单上:外蒙没了,满洲划进了苏俄的势力范围。
满洲后来保住了,但外蒙还是独立出去了。
上次就在菜单上的日本当然害怕这次又上了菜单。所以他们最早把习普川深圳会谈称为了“雅尔塔2.0”。
其实深圳三国会议不会成为雅尔塔第二。这次也就是拍个照,大家各取宣传上的所需:中国是“老子首次上桌”;俄国是“怎么说也还在桌上”;美国则是“桌上永远有俺”。
雅尔塔的时候,美英苏盟国是真有共同目标和利益:不管以后怎么样,但目前的纳粹和日本帝国必须消灭。
而现在,不要说中美俄没有任何的共同目标和利益,就连号称“友谊上不封顶”的中俄之间也没有什么共同目标和利益,比如普京最关心的是如何停止俄乌战争,但习近平肯定觉得干嘛不打个百年战争,起码要来个三十年战争嘛。
所以日本也还不需要担心在不在菜单上。
但这次的三方会谈最有意思的是告诉日本人,日本真的不是什么大国,而要看大国的脸色。
1
336
你才是傻逼 他们什么都用 然后发现还是得用opus
我观察到的一个危险现象。
拿我举例子,我是旗帜鲜明地支持全球人民消费中国deepseek、GLM、moonshot、qwen、minimax、混元、小米mimo这些大模型的,低强度和学生就应该闭眼买这些,经济实惠,为全球人类做贡献。
按道理是,全球市场份额很小的前提下,国产大模型吃国内央国企、体制内的订单, 应该能活得非常滋润。
但是我观察到的一个现象是,越是体制内、事业单位、高校、央国企这些不思进取的单位,里面混吃等死的程序员越对国产大模型有偏见,越坚信国产大模型是刷分作弊高手,只会刷benchmark,被GPT和claude吊打,甚至被gemini吊打。
哪怕我跟他们说,你们公司那点脑残傻逼业务破玩意儿,插入个国产LLM接入codex或者claude code甚至pi agent都绝对够用,拿个deepseek付费接进去就行,
他们都跟我直摆手,“你不懂,国产大模型真不行,全是垃圾,根本不能用,跟傻逼一样,不上GPT和claude根本不能干活儿”。
你说,这种人心中的成见,什么时候才能真正彻底打破。
1
192
Mstar retweeted
凌晨四点五十分,圣克拉拉园区。
黄仁勋拿着手提喇叭,站在声援他的员工面前:
“同志们,我来得太晚了。对不起同志们了。他们诋毁我、批评我,都是应该的。我这次来,不是请你们原谅的。
我想说的是,股价已经连跌三天了,你们的期权眼看要跌成废纸,不能再这样下去了。现在最重要的是,尽快结束这次静坐,回去等今晚的财报。
你们还年轻,来日方长,应该好好活着,看到人工智能改变世界的那一天。你们不像我,我已经老了,无所谓了。”



























