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

b11530: llama: keep the backend sampling graph static across ubatches (#30223)

October 9

The reserve builds n_outputs_max_per_seq sampling chains per sampler, while a decode built one per output row, so the graph changed its topology after the reserve and GGML_SCHED_NO_REALLOC builds…

Models & releases Chips & compute

Read at llama.cpp releases ↗

One letter a week, starting soon.

Latest

b11530: llama: keep the backend sampling graph static across ubatches (#30223)

Related

SiliconOpenAI and Synopsys partner to build "GPT-Synopsys" for autonomous chip design Tom's Hardware AI

ValleyAmazon’s $1B plan to combat data center backlash draws more backlash Ars Technica AI

SiliconHow NVIDIA GPUs Help Accelerate OpenAI’s GPT-6 Astra Ultrafast NVIDIA Blog

Voices[AINews] Gemini 4 Argon: GDM’s answer to Astra/Fable, with 1M output Latent Space

LabsIntroducing GPT-6.1 Sol OpenAI News

ToolsAnthropic integration with Modal brings scalable compute to Claude Science Modal