周报
2026-W34 周报精选：2.1.238；NVIDIA Vera Rubin and Blackw；Solving Agentic AI Fleet Cha
把中文全渠道内容（抖音 / B站 / 小红书 / 公众号 / X / 播客）采集进个人知识库的 13 个 AI Skill：图文存图、视频转文字稿、字幕优先免 GPU，附带知识库 MCP server。 ｜ Ingest Chinese content into your personal knowledge base — image/video routing, subtitle-first transcription, and a KB MCP server.
Added a keybindingFlavor setting: set it to "readline" to make Ctrl+W in the prompt delete back to the previous whitespace, as in Bash; the default ( "classic" ) is unchanged Plugin marketplaces: headersHelper on a url m
AI agents have expanded inference from single-turn interactions into multi-step workflows that reason, invoke tools, coordinate subagents, and carry growing...
AI factories are interconnected systems where fleet economics depend on how efficiently the entire stack converts power and capital into completed agent tasks....
GPT‑5.6 is now available in Kiro, helping developers plan, build, review, and test software with better price-performance.
Release: llm 0.33 My highlights from this release: Upgraded to the OpenAI Python library 3.x and switched the HTTP client dependency from httpx to httpx2 . #1608 , #1631 I shipped a quick 0.32.1 fix for this yesterday, b
NVIDIA Groq 3 LPX is the interactive AI inference accelerator for the NVIDIA Vera Rubin platform. At the core of the platform is NVIDIA Vera Rubin NVL72, the...
As AI agents become more capable and operate over longer horizons, building security and trust into the applications they power becomes increasingly important....
And this was a debug session from hell, enormously helped by an AI doing much of the grunt-work. I'd like to call it my tireless helper, but the AI several times stated flat out that this was impossible and unsolvable an
Models keep absorbing the harness into their weights — soon, it will be a harness for human attention rather than for the model.
The key skill required to make productive use of coding agents is being able to confidently instruct them on how to make changes and then confidently verify that those changes have been applied in the correct way. Someti
Stop Making TUIs Thomas Ptacek advocates for building real native user interfaces for even the smallest of personal tools, because coding agents have reduced the cost of getting a usable-enough GUI up and running to almo
A frontier language model is only one component of an AI agent. The surrounding agent system—often called a harness—determines how the model receives...
Google DeepMind partners with game studios to prototype breakthrough AI gameplay.
Prior to Fable, it felt silly to waste too much time improving your coding harness or context strategies. A new model would arrive at the same price (or cheaper!) and paper over most of your problems. But then Fable land
**Agent harnesses** are becoming a key optimization focus, with NVIDIA research showing traditional skill checks poorly predict agent usefulness and proposing a new metric called **"Skill Lift"**. Open-source implementat
Release: llm 0.32.1 Fresh installs of LLM stopped working the other day because the OpenAI Python library dropped its usage of httpx , and it turned out LLM depended on that library but only installed it via a transitive
**Z.ai** released the **GLM-5.3** open-weight model family, optimized for **agentic coding** and **cyber defense**, with impressive specs like **744B total / 40B active parameters**, **1M context window**, and a **239GB
**Stanford** is formalizing AI-native software engineering with a major curriculum overhaul replacing **85% of Fall 2025 material** to focus on **agent skills, context engineering, MCP portals, agent-ready codebase desig
Release: llm-openrouter 0.7 Now that this plugin is compatible with LLM 0.32 it can display the reasoning traces for LLMs available through OpenRouter. Updated for compatibility with LLM 0.32 . Models now use OpenRouter'
人负责关键判断，Agent 做分析、设计、编码和验证。用 BaoCut 远程转录把这套流程完整走一遍。
We ran 904 DeepSWE rollouts on GLM-5.3 and GPT-5.6 Sol. Sol leads pass@1 by 3.7 points; GLM-5.3 wins pass@4 at half the cost, and a GLM-first cascade hits 85.9%.
We ran 904 DeepSWE rollouts on GLM-5.3 and Claude Fable 5. A tie on pass@1, but GLM-5.3 wins pass@4 and costs 5.4x less: \$3.99 per rollout vs. \$21.63.
Did you think RSI stopped at model training?
The retrieval layer that helps AI systems navigate, read, and verify information inside even the most complex documents
With a fixed deadline and design resources committed elsewhere, Stampli used Codex and ChatGPT Work to compress weeks of launch production into days.
"If you apply enough compute to matter, you get this thing that looks like intelligence." Max Hodak on why AI models and brains seem to represent concepts using the same underlying geometry.
OpenAI reaffirms Zero Data Retention for eligible API customers and previews Private Safety Processing for advanced AI safety without compromising data privacy.
Simile’s CEO about his journey from the viral Generative Agents to creating 8 Billion Digital Twins of every living human... and why it’s gone from fun exploration to very serious business.
Introducing Intelligence Age, a new OpenAI blog exploring how transformative AI could reshape power, governance, the economy, and individual freedom.
A 48-minute video walkthrough of token sampling, watermark detection, and removal
Anthropic’s best AI model struggles to attract users as cheaper tools thrive A few interesting numbers in this FT story gathered from "people with knowledge of the matter": Anthropic's "annualized revenue" for July is up
NVIDIA Holoscan is a platform for building real-time AI applications at the edge, from medical imaging to robotics. HoloHub is its companion repository: a...
Longitudinal electronic health records (EHRs) document patients' sequences of clinical visits over time, preserving the temporal evolution of disease progression and care delivery. However, real longitudinal EHRs are dif
**OpenAI** announced benchmark results for its custom inference chip **Jalapeño**, showing **1.5–1.9×** better efficiency and **1.7–3.6×** lower latency compared to NVIDIA **GB200/GB300**. Deployment starts by year-end w
Vector sketches remain one of the most concise and immediate mediums for abstract human expression. However, generating high-quality vector strokes that exhibit human-like drawing styles remains an open challenge due to
Black-box VR and 3D applications are difficult to regression test because observable failures depend on where a tester moves, what objects are visible, and which views are captured. Manual exploratory testing can find su
**OpenAI** and **Anthropic** expanded their agent platforms with new desktop features, collaborative editing, and composable APIs like Skills and Files API. **OpenAI** rolled out memory and workflow features in the EEA,
ChatGPT search now uses the site:operator at scale Promptwatch is part of the emerging "GEO" space, for Generative Engine Optimization - the chatbot version of SEO, where companies offer tools and consulting to help your
AI agents are only as effective as the context they receive. Even with capable models and well-documented NVIDIA libraries, agents can spend extra steps finding...
Here’s how you can use Google Search tools to study for classes and standardized tests.
Replit introduces Free Mode, powered by GPT-5.6 Luna, so anyone can turn ideas into working software without worrying about token costs.
Differential acceleration of cyber, math, and AI
Modern vision-language models (VLMs) can support tasks such as visual question answering, captioning, and image-text reasoning. In practice, however, the data...
**Microduck**, a **25 cm open-source biped robot** from **Pollen Robotics** and **Hugging Face**, priced at **$399** and shipping before Christmas, features **15 actuators** and a rich sensor suite including camera, LiDA
LLM agents increasingly take on long-running tasks whose history grows far beyond a single model context window. Existing approaches compress earlier interactions or extract selected information into fixed memory represe
Large language models (LLMs) are increasingly used to simulate students at different mastery levels. These simulations can generate synthetic training data and stress-test tutoring systems. However, common prompt-based a
ChatGPT Ads is expanding to 31 European markets. Learn how advertisers can reach people as they explore, compare options, and make decisions.
Retrieval-Augmented Generation (RAG) has emerged as a powerful paradigm for combining large language models (LLMs) with external knowledge sources. However, RAG systems remain vulnerable to prompt injection attacks, whic
Atomistic simulation requires three things: knowledge of the science, compute-efficient implementation of simulations, and accessible interfaces to the...
Asana used OpenAI Codex to replace an outdated testing system in two weeks, completing work expected to take five years for about $12K.
OpenAI launches an initiative to strengthen democratic oversight of AI in national security, supporting government institutions with tools, training, and expertise.
OpenAI and CodeAI are partnering to help students build AI literacy, think critically about AI, and develop the skills to use and shape it responsibly.
OpenAI is strengthening monitoring, alignment, and security for frontier AI models. See how new safeguards are guiding the pace of model development.
LLM agents learn by interacting with environments, yet these environments are hand-built and static: blind to an agent's weaknesses, and quickly left behind as it improves. While recent environment generation methods att
ChatGPT for Teens helps teens learn, think critically, and use AI with confidence, with stronger built-in protections, healthy-use features, and additional controls for parents.
Software increasingly functions as part of the scientific instrument itself, making failures in scientific code capable of compromising not only program behavior but also the evidence underlying scientific conclusions. Y
Every lab CEO is on X now
Modern LLM agents are often improved by modifying prompts, tools, or workflows manually, while the executable scaffold surrounding the model---the harness---is typically treated as a fixed artifact after deployment. This
AI is reshaping cybersecurity for attackers and defenders alike. Learn how OpenAI is strengthening its defenses and what security teams can do now.
LLM coding agents issue Bash commands through interfaces that may serialize, wrap, and reparse model output. Matched execution scores alone cannot distinguish command-generation errors from failures introduced after gene
NVIDIA teams use ChatGPT Work to reduce manual tasks, connect fast-moving signals, and scale successful workflows globally.
the Memory crunch continues - Moore’s Law reversed to 2007 levels
We ran 904 DeepSWE rollouts on DeepSeek V4 Pro 0813 and GPT-5.6 Sol. Sol leads pass@1 by 10 points at 35x the cost; Pro wins pass@4, and a Pro-first cascade hits 83.0%.
OpenAI joins PORTS-Pike project, expanding community investment and supporting thousands of Southern Ohio jobs
**Ornith-1.5** launches as a new open-weight model family with **9B dense, 35B MoE, and 397B MoE** variants under **MIT license**, featuring quantized formats like **FP8, GGUF, MLX, and NVFP4** and showcasing end-to-end
Google Gemini and Pixel partner with five global football clubs to elevate the fan matchday experience through AI and Smartphone Technology.
OpenAI funds 14 independent projects exploring new AI policy ideas to expand economic opportunity and strengthen societal resilience in the Intelligence Age.
This article is from Making AI Work, MIT Technology Review’s limited-run newsletter examining how to apply LLMs across industries. To receive it in your inbox, sign up here. Chatbots took many schools by surprise upon th
Glean CEO Arvind Jain explains why model routing helps control AI costs for organizations, and how human feedback loops at scale improve its routing systems.
上月底，豆包与飞书合并。 上周，豆包推出了工作模式。 昨天，字节跳动正式发布了「豆包工作」。 字节这一系列动作快得不行，目标清晰：做一个面向生产力场景的独立 Agent 产品。 豆包给人的印象是适合聊天，但新出的「豆包工作」在办公场景到底能不能打？直接上案例。 「豆包工作」现教现学 先说说背景，昨天一个大学同学找我，想
We ran 904 DeepSWE rollouts on DeepSeek V4 Pro 0813 and Claude Fable 5. Fable leads pass@1 at 90x the cost; Pro wins pass@4, and a Pro-first cascade hits 82.7%.
Shadow traffic proves a candidate is operationally sound. It can't tell you if users like it better. Run the split at the endpoint instead of in your app code.
字节最近频频发力，扣子最近出了客户端，从官网看，主打也是办公场景！ 前几天测了豆包工作模式，扣子之前擅长做工作流，现在感觉彻底Agent。 https://www.coze.cn/