Pinned Tweet
2006 年開始拍人像,出過三本《制服.女孩》。現在用 AI 做寫真集,一本 36 張,每天在這裡發一張。追蹤我,一張都不會漏。
Portraits since 2006, three Uniform Girl books. Now 36-frame photobooks made with AI, one frame a day. Follow so you don't miss one.
1
1
231
拍這張,我會把黃色圓板收進前景,留一小塊藍在身後。你覺得那一點藍,讓畫面更安靜,還是更醒目?
I'd frame the yellow disc in front, leaving a little blue behind her. Does that blue feel calming or striking to you?
1
14
《讓房間保留一點藍》完整 36 張在這裡:
Leave a Little Blue, all 36 images:
swanky.github.io/photography…
9
在提示詞寫「不准合併 PR」,AI 說記住了;三小時後對話被壓縮,它還是合了。
Hermes 插件黑客松前 10 名,多數把這種規則搬到工具執行前的關口。我讀了十個專案的程式,每件畫一張圖:怎麼擋、哪裡擋不到。
swanky.github.io/technical/h…
18
客戶自研晶片這段我同意:NVIDIA 最大的對手是它自己的客戶。這會決定往後的 AI 服務多少錢、被誰綁住。會不會因此失敗,我沒那麼確定。
I agree that NVIDIA's biggest threat is its own customers. That will decide how much AI services cost, and who locks you in. Whether that means NVIDIA fails, I'm less sure.
Ok I do have a view on NVIDIA without Jensen.
I’ll tell you the real reason $NVDA will likely fail without him and why the frontier labs and hyperscalers are the biggest winners. (Elon has been in a really good mood recently. Have you noticed?)
The biggest threat to NVIDIA is its own customers, the frontier labs. NVIDIA has to become a frontier lab faster than frontier labs are becoming chip companies. And with Jensen out, they likely won't maintain the momentum to get there.
With recursive self-improvement (RSI), every frontier lab and hyperscaler must become a chip company or lose the AI race. It's that simple, and they've all laid the groundwork (Google TPUs, OpenAI and Meta building custom ASICs with Broadcom, Anthropic building an in-house silicon team, and SpaceX/xAI/Tesla building Terafab).
That's because software-only RSI has clear physical limits. Software can reduce how much computation and data movement a model needs, but it can't change the chip's physical bandwidth, memory capacity, or interconnect. Custom silicon gives labs another way to improve performance: designing the memory system, compute, and networking around their own workloads.
To get the real compounding effect of RSI, you have to co-design the model and the silicon so it becomes a two-way feedback loop: smarter models design better custom chips, and those chips train and run the next model faster. You can't do that on off-the-shelf NVIDIA GPUs because you get zero proprietary edge buying the same chip as everyone else, and getting your next-gen architecture into NVIDIA's silicon means sharing your roadmap with them years before tape-out (which they then sell to your rivals). Plus, AI now automates writing compilers and porting CUDA kernels to other chips, which kills the CUDA moat.
Jensen argued custom ASICs aren't more cost-efficient than GPUs because models change too fast for fixed hardware, and partners like Broadcom charge fat margins too. But hardware doesn't fall behind model changes when AI co-designs the next model and chip together instead of humans designing them separately (which Google and OpenAI are already doing). And labs aren't building custom silicon just to save on NVIDIA's 75%+ gross margins. Closing that hardware-software loop is the main goal, though the cost savings are still massive since RSI requires 24/7 internal inference where custom ASICs are way cheaper and more power-efficient than general-purpose GPUs.
That's why NVIDIA has been trying so hard to become a frontier open-weight lab, buying Groq (fast inference IP), Poolside + Nemotron (internal coding models), and Hugging Face (distribution). If closed labs win, enterprises just call their APIs running on the labs' own custom silicon and NVIDIA gets cut out. Subsidizing free open-weight models is the only way to make enterprises buy or rent NVIDIA GPUs to run the models themselves.
With Jensen gone, it's unclear if the Hugging Face deal even survives DOJ approval (can a GPU company own the whole stack?), and a non-founder is unlikely to take the massive risks needed to become a frontier lab: putting billions of capex into internal clusters and free open-weight models instead of just selling GPUs, while competing directly with its own biggest customers $GOOG, $META, $SPCX, $ANT, $MSFT
19
Swanky Hsiao | swanbear.eth retweeted
The Hermes Agent Plugin Hackathon winners are here. Thank you to the nearly 1,000 community members who submitted!
Agents scored every entry, then five community judges ranked the finalists to pick the top 10.
Winners below:
The Hermes Agent Plugin Hackathon starts now!
It runs for 7 days, with $20,000 in cash prizes and more.
Plugins are the most powerful way to extend Hermes Agent, and we want this hackathon to show what they can do. A plugin can change anything from the model Hermes runs on, where it stores its memory, where it runs commands, which messaging apps it works in and what Hermes Desktop shows. Plugins in the marketplace already let Hermes do things like control a phone or turn a photo into a printable 3D model.
Every kind of plugin is eligible, so build whatever functionality you want Hermes to have. We'll list accepted entries in the new Hermes Marketplace, where every Hermes user can install them.
Prizes
1st: $10,000 + neon sign + merch + $1,000 creds
2nd: $5,000 + neon sign + merch + $1,000 creds
3rd: $2,500 + neon sign + merch + $1,000 creds
4th: $1,500 + merch + $500 creds
5th: $1,000 + merch + $500 creds
Runners up (x5): merch + $250 creds
Submissions are due 5PM EST on Nov 11th. To enter:
1. Open a pull request that adds your plugin to the Hermes plugin catalog: hermes-agent.nousresearch.co…
2. Post a demo video on X and tag @NousResearch
3. Submit both links through the form: form.typeform.com/to/FwJLvOl…
Your entry is the plugin itself. Judges score it on usefulness, reliability, polish and creativity, and they favor plugins that do more than a prompt or a skill could. Keep the demo under 90 seconds and show the plugin working.
The developer guide shows how to build and test a plugin: hermes-agent.nousresearch.co…
When submissions close, we'll open a thread for every entry in the Nous Research Discord, so you can browse all the plugins and talk to the people who built them.
You can submit the form as soon as your pull request is open. The catalog checks run once a maintainer starts them, and we review entries every few days.
If you have any questions stop by the Nous Research Discord: discord.gg/nousresearch
GLHF!
Ruby on Rails creator DHH says the last 2% of building software is where he gets the most satisfaction:
"What we're seeing with a lot of vibecoding projects is they start out and they're kind of promising, and then someone loses interest... Let me just get something done in the first 80%, and then the last 20 is where things are a little janky, they're not quite right, they have some bugs, whatever, and they lose interest."
"For whatever reason, I am blessed with the entire dopamine payoff in the last 2%. I don't actually give a fuck about the first 98. I want the last two. The last two is where I get my satisfaction of building software."
"If you are of that type, that you just keep at it and keep at it and keep at it until it's literally perfect, this is an extremely addictive moment in history because there at least, for the time being, is still a place for someone to drive these agents."
"I've not yet found the correct loop or setup where you can just one-shot everything and the design comes out perfect and everything comes out perfect."
@dhh
Ruby on Rails creator David Heinemeier Hansson says AI is giving small teams a chance to compete with software ecosystems built by thousands of developers. In his view, as building software gets faster and cheaper, taste and craft become more important.
DHH joins a16z General Partners Martin Casado and Anish Acharya to explain why he's building Omarchy, an agent-first Linux distribution. They also cover why Linux is unusually well suited to agents, the importance of open models, the return of individual hackers shaping the future of computing, and more.
00:00 Intro
00:48 From Rails and Basecamp to Linux
03:58 Why AI agents make Linux the better OS
08:17 Can a third mobile platform happen?
16:02 Does taste still matter when AI writes code?
22:57 What counts as superintelligence?
25:58 Big AI labs vs. open source
31:50 DHH's plan to be bigger than macOS
38:24 Who to follow, and the AI culture war
44:35 The return of the pseudonymous hacker
47:33 90s demo culture and computers with character
YouTube: youtu.be/TtOoLanFzGw
@dhh @martin_casado @illscience
隱私錢包 Fluidkey,在 36 分鐘內被轉走至少 158 萬美元。錢是從它自己的轉發器出去的。官方已暫停服務,但還沒公布原因和損失金額。我把已核對過的數字、時間線,以及一般人現在該查什麼,寫在這篇 X 文章裡。
A privacy wallet called Fluidkey had at least $1.58 million leave user accounts in 36 minutes, through its own relayer. The company has paused its services and has not yet said how it happened. I wrote up the verified numbers, the timeline, and what ordinary users should check now.
1
133
拍這張,我會保留髮絲邊緣的光,讓海港退到背景。你喜歡望向鏡頭,還是看向畫面外的人像?
I'd keep the light on her hair and let the harbor recede. Do you prefer portraits looking at you or beyond the frame?
1
1
56
《重力休息的那一天》完整 36 張在這裡:
The Day Gravity Rested, all 36 images:
swanky.github.io/photography…
14
Swanky Hsiao | swanbear.eth retweeted
马斯克谈到老朋友黄仁勋都快哭了!!!
马斯克人在深圳,随特朗普代表团出席 APEC,隔着越洋视频连线DealBook。当聊起黄仁勋的时候,这个平时怼天怼地的男人,声音一下子软了下来。
画面里他低着头,说着说着眼睛微微闭了一下,看起来像是在忍着什么。开口第一句就是:“黄仁勋是我老朋友。”
听得人心里真不是滋味。他替老朋友辩护,承认黄仁勋“确实犯了一些错误”,但话锋一转,语气极其笃定:“我不认为他对美国不忠。”
“感觉就像在拍电影,像乔治·R·R·马丁的剧本里,你已经投入感情的一个主要角色,突然就没了。这相当令人震惊。”
真的,看视频那几秒钟,他声音明显软了下来,说着说着低下头,看起来像在生生忍着什么情绪。《权力的游戏》作者乔治·R·R·马丁最狠的地方是什么?就是奈德·史塔克,是血色婚礼,是你追了多少年、以为绝对有主角光环的人,毫无预兆,一集就没了。而在整个硅谷的剧本里,老黄就是那个主角啊。
马斯克顿了顿,说:“黄仁勋是我老朋友。你不会发现我公开说的和私下说的不一样。”
他承认黄仁勋“确实犯了一些错误”,但紧接着就替老友站了台:“我不认为他对美国不忠。”谈到老黄的离开,马斯克说:“在我看来,黄仁勋之于一支队伍,就像巅峰时期的迈克尔·乔丹……这就像在一波连胜里把迈克尔·乔丹踢出球队,对股东来说怎么看都不是好事。”
怎么替?真的,“没有人能真正取代黄仁勋。”
马斯克在镜头前细数老黄做过的事:一个人兼着好几个人的活,懂技术,还亲自抓供应链。在台湾和韩国,跟关键供应商蹲在路边一起吃街头食品,那是处了几十年的紧密交情。他的团队有多爱他?底下好多人因为对他的个人忠诚,硬是拒掉了别家更高薪的 offer。
马斯克透露,他们私下通过话,黄仁勋本人到现在都还处在震惊中。看到这句我鼻子真有点酸,那可是黄仁勋啊……半辈子心血,说没就没了。
现在请求老黄回归的请愿已经有 20 多万人签名,马斯克自己也签了。但英伟达董事长 Harvey Jones 在 Q3 电话会上已经把话死死堵住,公开表态董事会不打算请回他。马斯克也很无奈,说股东投票换掉董事会非常罕见,目前还是董事会掌控着公司。
但连线的最后,马斯克还是留了那句大实话——“他是历史上最有天赋的创业者 CEO 之一。谁会不想聘请黄仁勋?”
你追过的剧里,最爱的那个主角突然下线的那一集,你当时是什么感觉?
I think a lot of people aren't using Grok Bot to its full potential, because knowing how to automate things is actually quite a difficult skill to learn. I wrote this post to break down how I approach automating anything with Grok Bot!
Swanky Hsiao | swanbear.eth retweeted
Get on the Map. Get a Pack.
1. Link an external SOL wallet using the same Glyph account as your BAYC profile at useglyph.io
2. Go to boredapeyachtclub.com/member…
3. Head to your ‘IRL settings’ on your profile to check in and join the Club Map
Complete all 3 steps for a Mudwater pack.
Swanky Hsiao | swanbear.eth retweeted
Après Elon Musk, j’ai rassemblé toutes les interviews de Jensen Huang depuis 2003 sur un seul site.
283 vidéos. 241 heures. 1 867 citations. Et 14 de ses prédictions vérifiées une par une.
Tout se regarde directement sur le site, en 8 langues.
silicon-valley.tech
支持!
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
20
Opus 4.5 才一年前,Poteto 說感覺像上輩子。
我照 Matt Pocock 的 13 行時間軸,把每件事找回出處,再對照 Artificial Analysis 最新評比:頂尖模型已經擠在一起,卡關的地方換了五次。
swanky.github.io/technical/a…
37
Swanky Hsiao | swanbear.eth retweeted
OpenAI’s release of GPT-6.2 Astra and GPT-6.2 Sol delivers the #2 model in the Artificial Analysis Intelligence Index and most of the Cost per Task Pareto frontier
@OpenAI's GPT-6.2 Astra is priced at $8/$40, 20% below GPT-6 Astra, and cached input drops to $0.40 from $1.00 per million tokens. GPT-6.2 Sol pricing is unchanged from GPT-6.1 Sol, at $2/$10 per million input/output tokens.
Key takeaways:
➤ Second in Intelligence Index: OpenAI now trails only Anthropic’s Claude Fable 5.2 in the Artificial Analysis Intelligence Index. GPT-6.2 Astra jumps 5 points from GPT-6 Astra, overtaking Gemini 4 Carbon (60) and Claude Opus 5.5 (59) at their top effort settings. GPT-6.2 Sol gains 2 points at 57, placing it on par with GPT-6 Astra at ~15% of the Cost per Task.
➤ Occupies most of cost and output token frontiers: OpenAI models now occupy most of the Cost per Task Pareto frontier, sharing the frontier with only Anthropic and Meta. OpenAI also dominates the Output Token per Task Pareto Frontier, with Claude Fable 5.2 being the only model from a non-OpenAI lab appearing on the frontier.
➤ Strengths in science and knowledge: OpenAI performs strongly in both of our key science-focused evaluations, leading Terminal-Bench-Science 0.1, covering agentic scientific research workflows, and placing second only to Claude Fable 5.2 in CritPt v2 (hard subset), covering physics reasoning. GPT-6.2 Astra performs below only Claude Fable 5.2 in factual knowledge as measured by AA-Omniscience v1.1 Accuracy, and in HLE-Diamond, a reasoning and knowledge benchmark.
➤ Leading in 3D and robotics: GPT-6.2 Astra leads AA-3D at 1318, with Claude Fable 5.2 at 1271 and GPT-6.2 Sol at 1259. This benchmark measures how well models build interactive 3D experiences, and is rated by human graders. GPT-6.2 also leads AA-Robotics v0.1, our research preview measuring how well language models can understand, plan, and act in the real world. GPT-6.2 Astra and GPT-6.2 Sol scored 79 and 78 respectively, ahead of Claude Fable 5.2 at 75. Our team is hard at work adding more difficult tasks, as OpenAI and Anthropic models can now perform the majority of basic tasks in this initial research preview.
➤ No improvement in Coding Agent Index: GPT-6.2 Astra is flat with GPT-6 Astra, and GPT-6.2 Sol loses 3 points relative to GPT-6.1 Sol in the Artificial Analysis Coding Agent Index (Codex harness at max effort). For Sol, the drop is driven by a 14-point drop in SWE-Atlas-QnA-AA, where the model is more likely than its predecessor to omit key facts in its responses. Cost efficiency however does place several effort levels of GPT-6.2 Astra and Sol on the Pareto frontier for Cost per Task.





















