思想与社会2025 · 分享于 Hacker News
DeepSeek-V3 Technical Report
探索《DeepSeek-V3 Technical Report》的原文与 Hacker News 社区讨论，查看来源并加入阅读清单。
从论文，走向讨论
看看社区如何讨论这个想法。
来自 Hacker News 的原始讨论。不同的视角、疑问，以及值得继续思考的线索。
最多展示 100 条相关讨论。论文可能出现在评论中，并非原帖投稿内容。
在讨论中发现的引用
按 HN 条目顺序展示最近的 50 条引用来源。
评论 #48621663 评论 #48422249 评论 #48090147 评论 #48035455 评论 #47474952 评论 #47466183 评论 #47457357 评论 #47437432 评论 #47045914 评论 #46978916 评论 #46432740 评论 #45844859 评论 #45429456 评论 #44076189 评论 #43981176 评论 #43801660 评论 #43624206 评论 #43268531 评论 #43124583 评论 #43084252 评论 #43017358 评论 #42936877 评论 #42873275 评论 #42867506 评论 #42856385 评论 #42855604 评论 #42852187 评论 #42850611 评论 #42847126 评论 #42839808 评论 #42838514 评论 #42590438
Ynekofneko ·
Kimi K2 Thinking, a SOTA open-source trillion-parameter reasoning model
936 赞同427 评论
Ytalhof8 ·
Exposed DeepSeek database leaking sensitive information, including chat history
707 赞同474 评论
Yamrrs ·
The Illustrated DeepSeek-R1
578 赞同118 评论
YAnhTho_FR ·
Why DeepSeek had to be open source
525 赞同300 评论
Yrcarmo ·
Nvidia’s $589B DeepSeek rout
501 赞同1,073 评论
YCharlesW ·
Lossless LLM compression for efficient GPU inference via dynamic-length float
411 赞同117 评论
Ythm ·
Building a personal, private AI computer on a budget
384 赞同231 评论
Yt55 ·
Introduction to CUDA programming for Python developers
365 赞同95 评论
Yksec ·
Apple unveils new Mac Studio
348 赞同288 评论
Ypseudolus ·
Meta got caught gaming AI benchmarks
347 赞同161 评论
Ysemking ·
Berkeley Researchers Replicate DeepSeek R1's Core Tech for Just $30: A Small Mod
163 赞同66 评论
YLorenDB ·
Sam Altman said startups with $10M were 'hopeless' competing with OpenAI
142 赞同154 评论
Ybchelli ·
A path to O1 open source
133 赞同80 评论
Ysigna11 ·
DeepSeek-V3 Technical Report
132 赞同34 评论
Ymooreds ·
Attention Wasn't All We Needed
130 赞同24 评论
Yblumpy22 ·
DeepSeek’s founder is threatening US dominance in AI race
89 赞同77 评论
Ytimermore ·
USA restricts Swiss access to AI computer chips
73 赞同133 评论
Yashvardanian ·
A Gentle Introduction to CUDA PTX
59 赞同8 评论
Ylimoce ·
Step-Video-T2V: The Practice, Challenges, and Future of Video Foundation Model
41 赞同5 评论
Yprng2021 ·
Ask HN: Confused about how DeepSeek hurts Nvidia
36 赞同43 评论
Yamrrs ·
Accelerating Gemma 4: faster inference with multi-token prediction drafters
36 赞同7 评论
Ysxp ·
DeepSeek not as disruptive as claimed, firm has 50k GPUs and spent $1.6B
31 赞同13 评论
Ylayer8 ·
WD and Seagate confirm: Hard drives sold out for 2026
25 赞同0 评论
Ymeetpateltech ·
GLM-5: From Vibe Coding to Agentic Engineering
17 赞同6 评论
Yjnord ·
No, it doesn't cost Anthropic $5k per Claude Code user
4 赞同1 评论
Ypera ·
Super Micro Shares Plunge 25% After Co-Founder Charged in $2.5B Smuggling Plot
4 赞同0 评论
Ycylo ·
Local AI needs to be the norm
4 赞同1 评论
Y0xkato ·
How LLMs Work
4 赞同0 评论
Ydoener ·
DeepSeek-V3 Technical Report
3 赞同0 评论
Ycdrnsf ·
The Future of Software Development Is Software Developers
3 赞同0 评论
Ydevashish86 ·
Two Qwen3 models on one DGX Spark: the residency math
2 赞同0 评论
Ymatt_d ·
Mamba-3
1 赞同0 评论