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

Google Cloud AILabs

Best practices guide for customizing Gemini models via Reinforcement Learning (RL)

September 25

Reinforcement learning (RL) has been a keystone of modern LLM post-training, but it demands large training clusters and access to model internals that external customers can't have with proprietary…

Models & releases Research AI & Machine Learning Developers & Practitioners

Read at Google Cloud AI ↗

One letter a week, in your inbox.

Latest

Best practices guide for customizing Gemini models via Reinforcement Learning (RL)

Related

JapanEvaluating CLM-8B as a System 1 Decision Engine vs Jev and Kev Zenn (AI)

AsiaDeepSeek Details DSec Elastic Compute: Agentic-Training Sandboxes at ~3M/Day, 380K+ Concurrent Pandaily

AsiaDeepSeek details DSec sandbox infrastructure for agent training TechNode

GitHubGPT-6 Sol closes most of the alignment gap with Astra. It’s one-fifth the price. The New Stack

LabsTransformers now runs llama.cpp quants Hugging Face Blog

SiliconAccelerating Dropless MoE Training in JAX with NVIDIA Transformer Engine NVIDIA Technical Blog