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

vLLM BlogTools

Kimi K3 Performance Optimizations in vLLM: The Road to 2.8× Throughput

09.13

Kimi K3 serving optimizations across scheduling, KDA prefix caching, ReplaySSM state recovery, PD disaggregation and state offload, parallelism, MoE, and GPU kernels.

Open weights models performance kernels speculative-decoding

Read at vLLM Blog ↗

One letter a week, starting soon.

Latest

Kimi K3 Performance Optimizations in vLLM: The Road to 2.8× Throughput

Related

ToolsTiered KV Cache Offloading in vLLM vLLM Blog

ToolsGLM 5.3 Optimizations, Part 1: Hybrid HiSparse Offloading in vLLM vLLM Blog

VoicesOpen-Source AI & Open Models Reading List Interconnects

ToolsThe Open Source AI Stack Together AI

EuropeMistral raises €3B to make sovereign, open-weight AI the technology frontier Mistral AI

VoicesSED News: The NVIDIA-Hugging Face Deal, China’s Proxy Economy, the Open Weight Surge Software Engineering Daily