The name is the button.

An agent writes the code now, and a human thumb hovers over Accept all a hundred times a day. The honest question of this decade is what, exactly, you just accepted. So I read the diff.

Eight years as a backend engineer in Stockholm, back in Tokyo since 2023, and I still ship something small most nights, usually badly. My one rule: the only real review of a model is the thing you built with it. That is why there is no leaderboard on this site.

Around a hundred and forty sources come through — the labs' own posts, the papers, the tools, the runtimes and languages nobody is watching yet, the chip under the demo, and the newsletters people actually forward. What is moving gets counted so the week reads at a glance, and every few days I explain one mechanism in plain words.

A story here is a headline, a short excerpt, one image and the link to whoever wrote it. We never reproduce the post; the images belong to their authors. A post written in Japanese appears in Japanese, exactly as it was published, and the link goes to the original.

— Yuki Halvorsen, Tokyo

← Accept All Archive

llama.cpp releasesTools

b11724

10.19

CUDA: fix shared-memory WAR hazard in flash_attn_ext_vec ( #30573 ) CUDA: fix KQ write-after-read race in flash_attn_ext_vec Assisted-by: ChatGPT CUDA: apply review suggestions for KQ sync comments…

Chips & compute

Read at llama.cpp releases ↗

One letter a week, starting soon.

Latest

b11724

Related

Toolsb11720 llama.cpp releases

SiliconDatacenters 2027: The Year of Acceleration, +50GW in a Single Year SemiAnalysis

AsiaSG AI cloud provider Aolani eyes Malaysia data center deals Tech in Asia

AsiaShenzhen's International Quantum Academy Entangles Up to Six Phonon Modes on a 2D Chip Pandaily

ValleyNeocloud operator Nscale’s IPO plans could be at risk after Microsoft cancels UK data center project SiliconANGLE

AsiaNvidia-backed Nscale faces pushback on $30b IPO Tech in Asia