gpuix apps can now render in the terminal
create GUIs, use them as TUIs
- embed them in VSCode, Zed, Codex
- split GUIs into panes
- use all terminal power features for your GUIs
powered by the Kitty Graphics protocol
I turned @googlegemma EmbeddingGemma 2, an embedding model, into a speech-to-text model, and it turned out to be one of the fastest STT models you can run on device.
On a base M3 MacBook Air it transcribes an hour of audio in about 13 seconds: 288× real time.
The idea came in March with Gemini Embedding 2. It puts text, images, audio and video into one shared space, so a spoken word lands right next to the written word. If that's true, you should be able to read a transcript straight out of the embedding model. I couldnt tried it then, but last week EmbeddingGemma 2 came out as an open model, so I tried it.
Meta’s AI agent Muse can send emails and texts on your behalf. It’s also creating a map of everyone you know and love.
Every hour, Muse updates its map of your relationships with friends, family, romantic partners and coworkers, including “tensions and alliances.”
Hey @badlogicgames, I think I've found a better setup for @pidotdev. I had one old display panel, one old controller, hooked up to an ESP32, and now it's working. Waiting for the microphone to arrive
4/ It idles when it's off screen, so your games get the whole machine.
It's built for a controller, and mouse and touch work too. Runs on Linux, Windows and Android.
Game launchers shouldn't be a web browser in disguise.
So I built NOVA with Claude Opus. It's a native game launcher in C++ and OpenGL.
2.3 MB app
10 MB download
~145 MB RAM
0 Electron
Steam, GOG, Epic, emulators: one beautiful library.
nova.madari.media
@badlogicgames@krzyzanowskim@pidotdev I was able to reproduce this, I have a Ubuntu VM. Running tmux on it. I am using ghostty but using export TERM=xterm to run this.
we do this with overscroll-behavior: none
but we gate it behind a (pointer: fine) media query, so that it doesn’t impact touch devices. this means you can still pull to refresh on mobile
the right way to use model capabilities is not to ship 10x more features to prod
it's to spend more time understanding your users, trying experiments, building prototypes, learning about things you don't understand so that you can ship things that actually work
Pacing will last until chinese models become sota
Dario's direction is laudable - I just think we're past the singularity event horizon, and that slowing down is no longer a choice
Dude, why is RAM still so expensive? Like the price difference between UDIMM and RDIMM doesn't make sense. Don't make me start manufacturing RAM sticks.
I’m thinking about banning Claude code at Shopify until they change their mind and read AGENTS.md and .agents/skills etc.
Insisting on only reading CLAUDE.md sometimes leads to split brain problems when different team members use different tools. Just unnecessary.
@OpenAI@thsottiaux I was already charged $100 for my plan, but account is still showing Plus access. I had previously scheduled downgrade to Plus, and I’m now getting error:sub_schedule_conflict
“Updating plan with promotion when there is a schedule or future update scheduled..
2K Followers 7K FollowingPediatric pulmonologist and bioinformaticist, now in drug development. He/him. https://t.co/61YX7H46Pc @PioneeringMeds https://t.co/LqrSTRsx0C too
530 Followers 3K FollowingFather. Friend. Reader. Internet. Things with strings. Save the money, save the world. #Bitcoin podcast player https://t.co/KH1UOXZDq1 https://t.co/mBmhKrxqoD
719 Followers 7K FollowingNight comes so people can sleep like fish
in black water. Then day.
Some people pick up their tools.
Others become the making itself.
// Rumi
4K Followers 6K FollowingBuilding what comes next. 🦾 Eternal Optimist, UI Developer, Keen Observer of the Humanity. Former Lead Designer on AI for Harvard Business School 🏫
157K Followers 79 FollowingSane + 🌶️ takes in an insane AI world... AI capabilities researcher: co-created RLHF/ChatGPT @ @openai now trying to right the wrong 🤭 (ceo @typesafeai)
8K Followers 578 FollowingOmarchy Kids
Maker with --dangerously-skip-permissions
Solving a Dutch tax incentive with AI: https://t.co/e1Pf0FW4gC
Email me at [email protected]
2K Followers 19 FollowingAn open source design system that's fully customizable and agent ready. Try it out at https://t.co/rBHFBYS76h and https://t.co/OyHbxy4Jrl
66K Followers 84 FollowingSolve all disease.
Developing and applying frontier AI to unlock deeper scientific insights, faster breakthroughs, and life-changing medicines.
4.4M Followers 278 FollowingThe Indian Express brings to you latest news from India and the world, breaking round the clock. Get the App: https://t.co/IPRNSSmzII
44K Followers 3K FollowingChief Product Officer @ Docker.
Advisor/Board Member.
Former VP at Meta, Former VP at Google DeepMind, Former MSFT.
Opinions and em-dashes here are mine.
23K Followers 778 FollowingMade @elysiaJS. I cosplay, and code. No context, degen, programming, cosplay and occasional jirai-kei. Draw sometimes. See “highlight” for my dress up + cosplay
832K Followers 941 Followingchief ai officer @meta, founder meta superintelligence labs, founder @scale_ai. give everyone the adulthood they dreamed of in childhood
114K Followers 3K Followingvp of engineering @cognition. building @devinai.
previously: vp product @microsoft. vp of ai @vercel. founder of @turborepo (acquired by vercel).
3K Followers 544 FollowingI fix things other people designed broken. @Microsoft/@Azure, ex-@Sapo, @Vodafone, @Accenture. #apple #ai #linux #analytics @[email protected]
28K Followers 68 FollowingAI researcher & teacher @SCAI_ASU. Former President of @RealAAAI; Chair of @AAAS Sec T. Here to tweach #AI. YouTube Ch: https://t.co/4beUPOmf6y Bsky: rao2z