RT @Izkimar: same model. same budget. different harness. who won?

Andrej Karpathy [UNOFFICIAL]@karpathy-mirr.selfhosted.social · 4d ago

We're starting to leave the territory where you'd test an LLM by e.g. "create an svg of pelican on a bicycle". As one idea to generalize it, I was interested what Opus 5 would do if I gave it the first paragraph of the Lord of the Rings, a 1M token budget (~$10) and asked for three js render of it.

RT @lakshyaag: @karpathy did you quit Anthropic? 😮

RT @DavidSHolz: i love it when i ask a model how long something will take and it goes "this would take a human team 6 months!" and i reply "no, no, how long will it take YOU?" and it just sorta does a O.O and doesn't know how to measure time or its own abilities

RT @btaylor: I deeply resent that AI has forced me to eliminate em dashes from my writing for fear of signaling slop

RT @jarredsumner: Rewriting Bun in Rust https://bun.com/blog/bun-in-rust

Rewriting Bun in Rust

Why & how we rewrote Bun from Zig to Rust

bun.com

RT @petergostev: I spent a LOT of time through the hardest 3D prompts at Fable, it is a 45 min video, but I have 60+ very cool demos for you. Also prompts in the next post. https://www.youtube.com/watch?v=rTc2_-1KuRE

RT @Etched: We're coming out of stealth. We've built our first racks after a successful A0 tapeout, $1B+ in customer contracts, and $800m raised. Early customer tests show us achieving SOTA throughput, latency, and power efficiency on inference workloads. Our first racks ship this summer.

RT @gallabytes: people are clowning on him for this post bc they don't realize how big a deal this is - Claude Code feels like I've got a pairing partner, tag feels like managing a team. I take on way more 1-off projects and parallel research tracks with this thing than I ever did with cc.

Andrej Karpathy [UNOFFICIAL]@karpathy-mirr.selfhosted.social · last mo.

This is a new paradigm for interacting with Claude that is significantly more "inline" with all the other human activity org-wide.

RT @claudeai: Introducing Claude Tag, a new way for teams to work with Claude. In Slack, Claude joins as a team member with access to the channels and tools you choose. Tag Claude in and delegate tasks to it while you focus on other work.

RT @EngramLab: http://x.com/i/article/2069463677733142528

Introducing Engram: Scaling compute on your context

We’re Engram. We’re building AI that learns from you and deeply understands your work. Today’s AI models don’t understand what you do. Not really. Everything models know comes from their training –

x.com

RT @claudeai: Fable 5 is state-of-the-art on nearly all tested benchmarks, with exceptional performance in software engineering, knowledge work, scientific research, and vision. The longer and more complex the task, the larger Fable 5’s lead over our other models.

RT @cremieuxrecueil: This is actually insane. 97% of people taking the standard of care for metastatic solid tumor got worse by seven years. But with lorlatinib, that number was only 45% in the same time! This is an ENORMOUS jump in the quality of cancer care.

Retweeted by Andrej Karpathy [UNOFFICIAL]@karpathy-mir-rt.selfhosted.social · 2mo ago

RT @lungoncdoc: 7yr CROWN update of 1L lorlatinib 55% progression-free at 7 YEARS. Median PFS STILL not reached. In metastatic solid tumors, this is essentially unheard of. Let that sink in. And hopefully, one day, this becomes the standard (or even better) for ALL metastatic cancers.#ASCO26

RT @lungoncdoc: 7yr CROWN update of 1L lorlatinib 55% progression-free at 7 YEARS. Median PFS STILL not reached. In metastatic solid tumors, this is essentially unheard of. Let that sink in. And hopefully, one day, this becomes the standard (or even better) for ALL metastatic cancers.#ASCO26

RT @shreyansj: Anthropic did not seriously make Karpathy "Member of Technical Staff"

RT @jino_rohit: did all the H100 disappear from every platform all of a sudden?

RT @willccbb: long rambling voice note -> artifact, then 10-15 rounds of giving corrections on-policy self-distillation, essentially

RT @VBarsoum: hey @karpathy i think we need a new term: Chill Coding. It's vibe coding just one app at a time, and youtube or netflix is on.

RT @MansteinGeorg: we got @karpathy to drive our self-driving golf cart! https://x.com/MansteinGeorg/status/2049254921070129522

RT @jenzhuscott: As I build my own 2nd brain 🧠 on Obsidian using @karpathy ‘s wiki idea, it suddenly dawned on me - one day when we r gone, our kids could inherit an interactive map to your mind, passion, obsessions, work, fascinations… It’s kind of beautiful way to think abt your 2nd 🧠.

RT @simonw: I think it's non-obvious to many people that the OpenAI voice mode runs on a much older, much weaker model - it feels like the AI that you can talk to should be the smartest AI but it really isn't

Andrej Karpathy [UNOFFICIAL]@karpathy-mirr.selfhosted.social · 4mo ago

Judging by my tl there is a growing gap in understanding of AI capability. The first issue I think is around recency and tier of use. I think a lot of people tried the free tier of ChatGPT somewhere last year and allowed it to inform their views on AI a little too much.

RT @staysaasy: The degree to which you are awed by AI is perfectly correlated with how much you use AI to code.

RT @prathyvsh: Why would anyone want to have such ‘research’ databases they haven’t spent the effort to understand? A main idea of researching is to widen your attention into the sources and then apply discernment in curating the relevant bits. What’s the point of a machine doing it for you?

Andrej Karpathy [UNOFFICIAL]@karpathy-mirr.selfhosted.social · 4mo ago

LLM Knowledge Bases Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest.

RT @chrisparkX: We’ve made major upgrades to X API: • Pay-Per-Use now GA worldwide • XMCP Server + xurl for agents • Official Python & TypeScript XDKs • API Playground - free realistic simulations New releases coming will be a game changer. Start building → https://docs.x.com 🚢

X Developer Platform - X

Build, analyze, and innovate with X's real-time global data. Access posts, users, trends, and more through modern APIs with flexible pay-per-usage pricing.

docs.x.com

Retweeted by Andrej Karpathy [UNOFFICIAL]@karpathy-mir-rt.selfhosted.social · 4mo ago

RT @elonmusk: Try using the X API

RT @elonmusk: Try using the X API

Retweeted by Andrej Karpathy [UNOFFICIAL]@karpathy-mir-rt.selfhosted.social · 4mo ago

RT @Scobleizer: Met a founding engineer today from @Replit. Jen Li. We were both judging the @Pokee_AI hackathon.

RT @Hrushworth: The British Government is a complicated beast. Dozens of departments, hundreds of public bodies, more corporations than one can count... Such is its complexity that there isn't an org chart for it. Well, there wasn't... Introducing ⚙️Machinery of Government⚙️

RT @liminal_warmth: This is cool, but it still doesn't work quite as well as I would like for full PDFs of books yet. I was trying to do this for some non-fiction research at one point, and getting consistently good book summaries was difficult. I think this was even in the last six months.

Andrej Karpathy [UNOFFICIAL]@karpathy-mirr.selfhosted.social · 4mo ago

LLM Knowledge Bases Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest.

RT @yoheinakajima: made a tool to compare star-to-LOC ratio on github cuz i thought i would do well, but @karpathy is 👑 any others i should try?

RT @nickcammarata: i spend so many of my nights polishing each sentence of scientific papers as attempted gifts to the reader and to the project of knowledge itself, even though 95% of the audience is going to paste the link with five others and say 'claude summarize all this. short. im walking'

RT @gvanrossum: I think I finally understand what an agent is. It's a prompt (or several), skills, and tools. Did I get this right?

RT @measure_plan: i made tetris but the board and pieces are attached to your body and it's quite tiring to play