The World is Changing: AI For Creativity A few months ago, I sat in my office in Silicon Valley and watched as a tech founder showed me something extraordinary. On the screen was…
Hear me out: Agentic TPMs (Technical Program Managers.) I had this idea in Sydney while chatting with Martha McKeen at CBA. I think this winds up being the most immediate and dire…
I don't know, I think it really does damage the credibility of a movement with 60 year old voters in Ohio when it's hosting Slutcon and asking questions about sex with dead cows o…
1. Post 1 of 6 I'm realizing from the flywheel thread from @tobi that many people don't realize that small models can actually outperform larger models on a specific task - @MPara…
Btw we open sourced the core infra piece that makes these self improving loops possible. https://tangleml.com Quoted post by tobi lutke (@tobi) Training tiny models for special pu…
SpaceXAI engineer, Lauren Tan: "GrokBot is the most powerful agentic tool we have ever shipped, but only 1% of users use it correctly| at SpaceXAI, I'm running a team of 15+ GrokB…
Investors like Paul want the immigration debate to be strictly polar because it is the only way their traitorous ideas might seem reasonable. They paint one side as racist, backwa…
Regret the tone of my post on data centers yesterday. What I should have said: There were reasonable concerns about data centers 18ish months ago: water, taxes, jobs, electricity…
Change your parents' router DNS to 1.1.1.2 and 1.0.0.2 Cloudflare blocks malware and phishing at the network level Free antivirus that actually works. Nothing to install/update/di…
To clarify, I think the difference between the US and any European country, from a tech company-building perspective, is as follows: 1️⃣ The US runs a huge trade deficit with the…
1. Post 1 of 14 The dollar's share of FX reserves is at a two-decade low. Treasury convenience yields have turned negative. End of dollar dominance? No — but financial innovation…
1. Post 1 of 5 @astridwilde1 you can get global shutter cameras to do more fps by simply scanning less rows. i am writing my own drivers 2. Post 2 of 5 60fps depth: ) these are pi…
Post by となりのトトノ🏯Local LLM | Tonoken3 on X This is a major win. The new version scores a perfect 9/9 on xhigh — the runaway thinking that was making the old version fail 1 out of 9…
1. Post 1 of 2 🚨 EX-INSIDER BOMBSHELL: WE CREATED THIS EUROPE ON PURPOSE - CIA BOUGHT STOLTENBERG, RUTTE AND ENTIRE GOVERNMENTS I have never heard the truth stated so directly. Co…
I once taught an image processing course at Tesla. Now I am sharing it with you here—the journey of photon-to-numeric, narrated by Grok @bot. Video thumbnail from X post Video thu…
1. Post 1 of 10 The good news: we got much better at running The Sort. The bad news: we got much better at running The Sort. Quoted post by Crémieux (@cremieuxrecueil) The dollar…
Must Read: “Terra Incognita: The Economics of a Shrinking World.” Quoted post by Jesús Fernández-Villaverde (@JesusFerna7026) Last week I posted a new paper with Patrick Norrick:…
Thoughts About Scaling Law Scaling, but not only of parameters. Every model release now ends with the same question: how many parameters? It isn't a question that can be answered…
1. Post 1 of 2 Let's make a concrete plan to 'cure every disease in 10y' and see how far we could get. Our assumptions will be ~infinite IQ 200+ 'thinkers', automated labs by ~203…
A few reasons that I can see. Chips are less efficient in China, therefore China draws more power for every unit of effective compute, shrinking the electricity gap relative to th…
If LLM-as-a-verifier works as well as my last post suggests, this is going to be massive for local AI. Like the dev pointed out, pairing a large model with a small model lets you…
https://jax-ml.github.io/scaling-book/ read it now, thank me later Quoted post by Archie Sengupta (@archiexzzz) learn about inference engineering as fast and deep as you can Open…
1. Post 1 of 2 If you have a 16gb gpu this is for you, I even have a command line to you use to fit 128k context on this. 10gb qwen 3.8 27b that can code just fine. https://huggin…
1. Post 1 of 2 In case anyone wants a Grafana dashboard to monitor their DGX Cluster, I uploaded mine here https://github.com/RodriMora/dgx-spark-grafana-dashboard Image from X po…
1. Post 1 of 2 Running a full desktop environment in a microVM, and accessing it remotely over LAN (Omarchy, ping @DHH and @vaxryy 😊) QVM now supports ultra low latency desktop st…
omp has a very interesting eval/repl tool design. it runs a persistent ipython kernel pinned across session. this means imports, vars, open files survive even across subagents. th…
I would like to know what @bryan\johnson's girlfriend thinks about how economists on Twitter reacted to the SEP article on capitalism.
1. Post 1 of 2 1/2 Thanks Gavin for an especially thoughtful exchange. I don't usually spend much time on social media but I wanted to engage here because it really brings out the…
1. Post 1 of 2 vinge describes 4 singularities, and we are obviously in the first where machines alone are achieving superintelligence through rapid increasingly self propelled it…
1. Post 1 of 2 Ok. Becky just brought me another gin and tonic and is watching our daughter. So: I think the framing of cyber as defense dominant and bio as offense dominant is se…
1. Post 1 of 2 今天 DeepSeek 发布了他们的 Harness。如同当年 R1 的创新带给业界的震动一样，这一次他们又选择了一条和其他人完全不同的路。当大家都在做 Agent Harness 的时候，DeepSeek 却基于一句很特别的理念来设计自己的系统： "框架即插件，插件即框架。" Everything is a Plugin。模…
Follow my simple reasoning here. Large sparse LLMs are more powerful then dense one \while\ being faster and thus more energy efficient. Dense models are a need that is artificial…
GPT-5.6 Sol High was able to follow Fable's plan and produce a great version too. First stab was 30% slower, but one follow-up prompt got it to parity. Per-token cost: $43. In con…
Had GPT Sol do an optimization run on this. It immediately made it 63% faster. So now ttfx is 14x faster than the original tte. https://github.com/omacom-io/ttfx/pull/2 Quoted pos…
Fast H3 implementation for Metal. Enjoy, modify, and so forth: https://github.com/antirez/h3.c Contains code from @liuliu which is welcomed in taking back whatever parts he likes…
Is there really still no reasonable process to set up this damn trusted publishing on npm for new packages without having agent go to the settings pages in Chrome with my help?
Today we're also opening the weights for Muse Glimmer, a great 30B parameter dense model that can run locally. Soon we'll also release the weights for Muse Spark 1.2, our latest f…
DwarfStar with DFlash speculative decoding now can do the DeepSeek v4 Flash inference \much\ faster both when used with Metal and the DGX Spark.
We won’t be able to avoid Chat Control and government/corporate surveillance in the future. But we can render that surveillance worthless. A censor’s database is only valuable if…
DeepSeek Vision has to improve a lot before it's usable Image from X post Image from X post Image from X post Image from X post
Post by となりのトトノ🏯Local LLM | Tonoken3 on X 大事だったのは要素研究。これに尽きる。ファイルは資産。捨てちゃだめだった SSDのRAIDは正義。NASも正義 板状のSSDはすぐに壊れる前提に立ったほうがいい。汝OptaneSSDを求めよ
1. Post 1 of 2 This finally stopped Claude's crazy pursuit of talking more and more unintelligible every passing day \ Save in memory to always use ASD-STE100 Simplified Technical…
What. 24GB VRAM you now have 10M context fast model that hit 61% on terminal-bench-2.1 This will be incredible Quoted post by Pokee AI (@Pokee\AI) Releasing Pokee-Isaac 28B — the…
1. Post 1 of 11 Trip report: the experiment was a RESOUNDING success! Using my toddler as experimental bait, we walked through swamp and marshland, with a ton of still rain puddle…
I’m about as AI pilled as can be but I’m not sure what to make of this. I think it’s time for us to look at value created in relation to the true serving cost with some margin on…
two weeks ago i went on @swyx's pod and said some things that i... should not have said. a lot has happened since then, i owe you all an apology. i'm sorry that i was right about…
Add this to ur Codex config, makes it so that Codex will use the multiple choice tool outside plan mode. Makes it feel 100x better Image from X post Image from X post
After accepting for years that GMail is blinding me I finally asked my agent and it installed https://darkreader.org for me.
1. Post 1 of 2 I have long felt the diminishing effectiveness of LLM-as-a-judge for agents, and one day over coffee @mitch\troy gave me a rant that helped clarify why. The ground…
Shift heavy furniture on your own without hurting your back Video thumbnail from X post Video thumbnail from X post Watch video
The most reliably predictable trend of the next 100 years: every year, humanity will use significantly more computing power than the previous year. Someone should start an ETF bas…
Library as of 1 Oct 2026