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

MAGI-2 Preview on vLLM-Omni: Bringing a 114B Audio-Video MoE DiT to Moore Threads MUSA with torchada

10.12

vLLM-Omni runs Sand AI's MAGI-2 Preview audio-video generation model on eight Moore Threads MTT S5000 GPUs through torchada's CUDA compatibility layer, then optimizes MoE, mHC, attention…

Models & releases Open weights Chips & compute performance hardware multimodal vllm-omni

Read at vLLM Blog ↗

One letter a week, starting soon.

Latest

MAGI-2 Preview on vLLM-Omni: Bringing a 114B Audio-Video MoE DiT to Moore Threads MUSA with torchada

Related

ToolsNvidia reportedly agrees to purchase Hugging Face for $13B Weights & Biases

ToolsImprove TTFT for Multi-Turn Agentic Workloads with Bi-Directional KV Transfer in vLLM vLLM Blog

ToolsvLLM Support for NVIDIA Vera Rubin NVL72: 7.8x Throughput over GB200 NVL72 vLLM Blog

ToolsDeepSeek-V4.1-Flash on vLLM: 5x Agentic Throughput Since Day 0 vLLM Blog

ToolsAnnouncing vllm-metal: Concurrent Serving on Apple Silicon vLLM Blog

ToolsSplash Engine - the fastest local Qwen3.8 on Apple Silicon LM Studio Blog