今日必看

Build more responsive apps with the Decisions API

Fast, focused, low-cost decisions for any application.

Global default for Copilot features generally available

In September, we announced a global default policy for generally available GitHub Copilot features and supported client capabilities on Copilot Business and Copilot Enterprise plans. Starting today, we’re gradually rolling out enforcement of the policy through October 26, so it m

Meta API endpoints updated for Actions blob storage

We’re adding blob storage endpoints for GitHub Actions on github.com as part of our ongoing work to improve capacity and reliability. The meta API now includes these new endpoints. If your organization uses a network allowlist to access GitHub Actions blob storage, review your ne

OlmoEarth v1.3: State-of-the-art embeddings for the whole planet

What FDE means at OpenAI

What forward deployed engineering means at OpenAI

Running Frontier-Class AI Models Across 8 NVIDIA DGX Spark Nodes

AI developers aren’t just running a single model anymore—they’re orchestrating multiple models in parallel, chaining inference calls, and experimenting with frontier-class architectures that demand hundreds of gigabytes of memory. NVIDIA DGX Spark was built to bring AI developmen

Beyond tree cover: mapping the world’s forest types

A global map categorizing every 10-meter patch of land on Earth into distinct forest types.

Architecting Governed, Self-Evolving AI Agents on Google Cloud

Learn how Project Chrysalis combines Google Cloud's Gemini Enterprise Agent Platform, Agent Development Kit (ADK), BigQuery, and a four-gate defense-in-depth security stack (Model Armor, Cloud KMS, Binary Authorization, and OPA) to build governed, self-evolving AI agents that rem

Ensemble forecasting with IBM Granite time-series models

Donating Agent Skills to the Agentic AI Foundation

Donating Agent Skills to the Agentic AI Foundation News today from Anthropic: We’re committed to keeping Agent Skills open, community-driven, and vendor-neutral. Today, we are donating the standard to the Agentic AI Foundation (AAIF) under the Linux Foundation, which we co-founde

A rubric landscape for evaluating clinical reasoning in large language models: what exists, what is missing, and what needs to be combined

arXiv:2610.01938v2 Announce Type: replace-cross Abstract: Exam-style accuracy does not establish whether large language models (LLMs) reason well over clinical records. We define clinical reasoning as integrating and updating evidence across time and sources to form, revise and j

Agent Behavior Maps: Locating Agents by What They Would Do, Without Running Them

arXiv:2610.30014v1 Announce Type: cross Abstract: LLM agents with similar success rates can follow different action sequences. An Agent Behavior Map represents LLM-harness combinations using centered action log-likelihoods on shared reference trajectories, without requiring new e

Anthropic 指南：何时把子智能体合并进主 Agent 循环

Anthropic Applied AI 团队发布指南，讲解何时把 orchestrator 下的路由器、规划器和领域子智能体合并进单个主 Agent 循环。文中对比显示，带 skills 的单智能体在质量上常优于一提示全包和子智能体设计，且成本和延迟更低，一家客户改用单循环后运行时间约快 2x。

Why AI Compute Comes in Rectangles — Erik Bernhardsson, Modal CEO

Warm welcome to Basil Chatha , our newest cohost from the Forward Deployed side of the AI Engineering world! We’ll be ramping up a new show with him and couldn’t pass up an interview with Modal’s CEO Eric on the occasion of their monster Series D! Two years ago, we spoke with Eri

把 Muse 装进 ESP-Mosaico：做一个没用的东西，证明 Vibe Coding 也在硬件领域快速成长

Matrix首页推荐Matrix是少数派的写作社区，我们主张分享真实的产品体验，有实用价值的经验与思考。我们会不定期挑选Matrix最优质的文章，展示来自用户的最真实的体验和观点。文章代表作者个人观点 ... 查看全文

GitHub 通过 AI 辅助重构，将 Copilot 运行时迁移至 Rust

BAND for Developers

Coding agents, meet teamwork Discussion | Link

You are the Team — Anekha Sokhal, Moshi

Tickets and applications now live for AIE CODE - the last flagship AIE of the year. When Anekha Sokhal traded energy derivatives, staying on top of the market meant tracking more information than any person could read. Now, as CEO of Moshi, she is building the tool she wished she

Voxtral 3

Voxtral 3: Mistral's voice agent SDK for real-time speech in 50+ languages, accurate transcription, emotional TTS, deep tool use.

豆包工作上线 IGA Pages 插件，一句话让网页变得可分享、可共创

MiniMax Code 桌面端推出 Mini App 长运行应用功能

MiniMax 在 MiniMax Code 桌面应用（macOS / Windows）本地会话中推出 Mini App，一种带前端和后端的长运行应用，以插件形式安装并与 Agent 双向调用，让 Agent 的工作停留在常驻页面而非散落各对话中。

中国最新款机器人，把外国人看呆了！但关注点却歪了……

📌 一句话摘要 宇树科技发布的'功夫传奇卫士'H2-X 机器人，以其高超的武术动作和超强续航能力吸引了全球关注，但最令人惊讶的是它在视频中展示的'抬车'能力，引发了关于机器人是否真的能独立完成这一壮举的广泛讨论。 📝 详细摘要 宇树科技的 H2-X 机器人凭借其出色的武术表演和长达 8 小时的续航能力，在国际社交媒体上引起了轰动。然而，视频中一个看似简单的'抬车'动作却成为了讨论的焦点，许多人质疑这是否是真实的机器人能力，还是仅仅是视觉效果。尽管宇树标注了'道具实拍，仅供参考'，但这一场景仍然激发了人们对机器人未来在救援等领域的潜力的想象。 💡 主要观点

Claude’s personal agent

Hi folks, I’ve been working on a way to get my X bookmarks grouped in the right places for the newsletter. If you didn’t know, Keshav , who’s been working with me for many years, is leaving us at the end of the year to dive further into the exciting world of AI. He used to do a l

Integrated embedding on the Documents API

Integrated embedding on the Documents API

30 Months on the Exponential (in Charts)

My second book on AI, Co-Existence, is out today (and you can get it here ), 30 months after the release of Co-Intelligence in early 2024. Looking back across the gap between these books is a good way to get a sense of AI progress. And, thanks to AI, I can describe this progress

押注记忆原生，MemT 1.0 破局养成系 AI 下半场

用 AI 最让人崩溃的瞬间，莫过于你花半天纠正它把任务做对，换个新会话，它转头又踩进同一个坑里。模型做过了事，却没变成它的成长。跑通单次任务之外，它真正缺的是把反馈转化为经验，并在之后主动调用。 这种割裂在 Meta Muse、OpenAI Dots 与 xAI Grok Bot 掀起的私人助理热潮中格外明显：主流的外挂检索最多算翻查静态备忘录，没法把调教过程沉淀为行事默契，而这恰恰是长期助理最核心的门槛。 今天，记忆张量正式发布并开源 MemT 1.0 记忆原生大模型。 它和外挂方案最大的不同，不只是让记忆脱离了外挂模块，内化进底层架构，更在于模型通过

Mercury 3 🤖, MiniMax Flash 3.1 ⚡, Exa 3 🔎

NotebookLM's Video Overview System Prompt: The Full Creation Workflow

The full NotebookLM Video Overview creation skill, with a brief introduction to its context gathering, review cards, and video generation workflow.