總是在那酸Deepseek的進來看看

Marc Andreessen: Deepseek R1 is one of the most amazing and impressive breakthroughs I’ve ever seen — and as open source, a profound gift to the world.

 

Nature:

https://www.nature.com/articles/d41586-025-00229-6

China’s cheap, open AI model DeepSeek thrills scientists DeepSeek-R1 performs reasoning tasks at the same level as OpenAI’s o1 — and is open for researchers to examine.

 

Grok:

"Deepseek R1 has achieved performance comparable to OpenAI-o1 in technical domains like coding, math, and reasoning.

It uses pure reinforcement learning, marking it as the first open research to validate that reasoning capabilities of LLMs can be incentivized without supervised fine-tuning.

The model is fully open source, allowing global access for examination, modification, and further development.

Deepseek R1 is notably efficient, with an architecture of 671 billion parameters where only 37 billion are active during operation.

It has rapidly gained adoption among top U.S. university researchers and companies, signaling a shift in AI innovation towards China.

Deepseek R1's development and release coincide with discussions on China's growing influence in tech and AI, challenging the status quo."

所有跟帖: 

就是說這個確實是比美國的大語言模型要好不少了 -芳華萬裏- 給 芳華萬裏 發送悄悄話 (0 bytes) () 01/24/2025 postreply 12:00:30

穀歌前總裁 Eric Schmidt 是個非常自信 assertive 的人,最近在一次采訪中不無焦慮 -遼闊的海- 給 遼闊的海 發送悄悄話 遼闊的海 的博客首頁 (75 bytes) () 01/24/2025 postreply 12:06:34

Eric Schmidt 來大千股壇, 都會被說成是歪果仁了。 -yangcen- 給 yangcen 發送悄悄話 yangcen 的博客首頁 (82 bytes) () 01/24/2025 postreply 12:30:00

哈哈,歪果仁的定義變了 -遼闊的海- 給 遼闊的海 發送悄悄話 遼闊的海 的博客首頁 (0 bytes) () 01/24/2025 postreply 12:33:59

算法上中國追上來了。算力中國還不夠。中國問題是數據,AI的基礎是大量的“真實”的數據。 -HBW- 給 HBW 發送悄悄話 (0 bytes) () 01/24/2025 postreply 12:09:36

明白人。 -TalkToMi- 給 TalkToMi 發送悄悄話 (0 bytes) () 01/24/2025 postreply 12:18:12

中國人聰明,如果用對地方,那絕對行 -Pilot007- 給 Pilot007 發送悄悄話 (0 bytes) () 01/24/2025 postreply 12:10:55

關鍵是它還是跟在美國屁股後頭跑,並沒有真正的創新,稍微快一點說明不了太多,而且回答也是錯誤連篇 -老北京炸醬麵- 給 老北京炸醬麵 發送悄悄話 (0 bytes) () 01/24/2025 postreply 12:18:00

我敢打賭你不僅不是這行的,而且連用都沒用過 -dancingpig- 給 dancingpig 發送悄悄話 (0 bytes) () 01/24/2025 postreply 12:51:01

比較進去看,這個不好作假 -dancingpig- 給 dancingpig 發送悄悄話 (193 bytes) () 01/24/2025 postreply 12:53:08

美國人從來都是重視對手,也誇大對手,這就對了 -老北京炸醬麵- 給 老北京炸醬麵 發送悄悄話 (0 bytes) () 01/24/2025 postreply 12:19:00

我測試的三個免費版本裏麵(germini, Chatgpt,deepseek)deepseek結果最好,而且好很多 -ocliving2005_4ever- 給 ocliving2005_4ever 發送悄悄話 (0 bytes) () 01/24/2025 postreply 12:42:55

噓,小聲點,不要驚動了山西佬鎮江佬 -Helen1817- 給 Helen1817 發送悄悄話 Helen1817 的博客首頁 (230 bytes) () 01/24/2025 postreply 12:44:34

瞞不住了,Reddit 上老美都在討論 Deepseek -遼闊的海- 給 遼闊的海 發送悄悄話 遼闊的海 的博客首頁 (0 bytes) () 01/24/2025 postreply 12:51:20

請您先登陸,再發跟帖!