arrow_back
Threads
Threads 1257
Threads
For a decade, we've made models wider and deeper—but we've barely changed how layers *talk* to each
Threads
if you had to restart your career from zero tomorrow, what's the first thing you'd do?
Threads
/handoff might be my new favourite skill https://t.co/VZ4nTjqWdx
Threads
Today we’re releasing EMO, a new mixture-of-experts (MoE) model trained so modular structure emerges
Threads
MAI-Thinking-1 is out!
Threads
Introducing MIRA, a multiplayer world model, a dream of Rocket League
Threads
@arnie_hacker: 👀
Threads
3 Months Fully funded AI Program in Singapore.
Threads
The Danish Foundation Models (DFM) project is adapting our modular FlexOlmo architecture into a ligh
Threads
A tiny recursive model beats LLMs thousands of times its size on ARC-AGI-style puzzles.
Threads
Foundations & Frontiers of AI
Threads
Companies are hiring remotely in 2026 and paying $90K–$210K+.
Threads
发现一个巨牛的书 The Hitchhiker’s Guide to Agentic AI,覆盖 Agentic AI 全貌
Threads
现在顶级 AI 实验室的入场券,早就不只是有学术光环了!
Threads
last night Boris Cherny (head of Claude Code) noticed his CI was slow. he gave Claude one sentence a
Threads
i have been a big article and essay enjoyer for 7 years.
Threads
https://t.co/mFvSsFIzlm
Threads
https://t.co/OMBzNQETKM
Threads
holy crap, Lettera will be the de facto "markdown ide" within a year
Threads
deep-ml is seriously awesome. It was great a year ago, when all it had was a list of questions to wo
Threads
Harness EngineeringからRSIへ
Threads
One of our core beliefs at Activate is that India's AI opportunity won't come from simply applying f
Threads
i don't know what a world model is but this looks interesting
Threads
I think there are books that teach you how to build software in 2026, they’re just of a different fl
auto_stories
Select a bookmark to read
Click any item from the list