NVIDIA DGX Spark 64GB Gives Developers More Ways to Build and Scale Local AI
NVIDIA 宣布 DGX Spark 推出 64GB 统一内存新配置，10 月 23 日起由 Acer、ASUS、Dell、Gigabyte、HP 和 MSI 发售，起步价 $4,999，支持最高 1000 亿参数模型在端侧运行。
Product Updates News
Latest reporting, research and product updates filed under Product Updates.
Latest stories
20 stories
A quiet landscape, gently in motion. 🌿💧❄️
A quiet landscape, gently in motion. 🌿💧❄️With 𝗗𝘆𝗻𝗮𝗺𝗶𝗰 𝗗𝗲𝘀𝗶𝗴𝗻 powered by 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮 𝟲.𝟴 𝗙𝗹𝗮𝘀𝗵 𝗣𝗿𝗲𝘃𝗶𝗲𝘄, Xiaohongshu creator @Muzlz8888 (Mr.木子李子) sets the sand swirling, the stream flowing, the ducks moving and the snow drifting — while keeping the buildings still.From a static image to a scene with rhythm. ✨
New feature sneak peek.👀
New feature sneak peek.👀Can you imagine The Thinker in snow boots? Meshy can make it happen.Select a specific area and type what you want to change. 3D creation is about to get easier and more fun.Meshy Edit. Stay tuned.
Turbo Haskell
Exactly a week ago (as a joke), I started writing THC, my “Turbo Haskell compiler,” while on vacation visiting Bartosz Milewski.It has grown a tiny bit since then.THC now implements every one of GHC 9.14.1’s prim-ops and provides a JIT for GHC Core that runs Haskell on the JVM. It uses the approach for running typed functional languages I developed several years ago in Cadenza (talk), using Truffle and GraalVM.GHC still handles parsing, typechecking, desugaring, and Core optimization. THC takes
Turn Any Character Into a Talking Presenter with PixVerse
Turn Any Character Into a Talking Presenter with PixVerseAdd the new PixVerse plugin and choose your character.They’ll speak, react, and gesture naturally on camera while PixVerse generates the clips for you.
AI music maker Suno now generates spoken words
Suno is branching out from the world of AI music, launching a new feature that generates spoken voices based on scripts or prompted descriptions. Speech is now available in public beta across Suno’s web and mobile platforms, and allows you to simultaneously generate voiceovers and background music to accompany them.“Music will always be at the heart of Suno and what we build. At the same time, our vision has always extended to other forms of human expression,” Suno chief product officer, Jack Br
3 Questions: A new resource to empower young entrepreneurs
The book “Disciplined Entrepreneurship” by Bill Aulet, managing director of the Martin Trust Center for MIT Entrepreneurship and the Ethernet Inventors Professor of the Practice at the MIT Sloan School of Management, walks readers through the 24 steps of starting a venture. With more than half a million copies sold, the approach has proven remarkably effective: MIT students who use the framework in the delta v startup accelerator program have a 61 percent survival/acquisition rate and have colle
Alpha Changelog - 10/1/26
ICYMI – last week, we launched a few features on alpha.midjourney.com that we're really excited about: style previews in the sidebar so you can see what you’ll get before you generate, larger images in create, and setting default parameters in a folder. Mostly focused on bugs and cleanup this week, as we get ready to launch some new collaborative tools next week. As always, the goal is making the site easier to use without taking control away. A lot this week came from #alpha-ideas-and-bugs and
Now in Grok Build: A new Agent Dashboard. All your agents on one screen.
Now in Grok Build: A new Agent Dashboard. All your agents on one screen.Try it with /dashboard
Pi Durable
Today Earendil and the Pi community shipped Pi 1.0. This reflects our belief that after countless hours of hardening, maintenance, and active development, Pi is now a solid foundation on which to build. Pi also continues to evolve. Together with Pi 1.0, we are shipping an experimental new package called Pi Durable. Pi Durable was built specifically for long-running, durable, and malleable agents that can run anywhere. We would like you to join in the fun and help us make it the best durable harn
We’re open sourcing a state of the art multimodal Decision Model, pplx-decider-27b, and are offering it in a new Decisions API at 4 cents...
We’re open sourcing a state of the art multimodal Decision Model, pplx-decider-27b, and are offering it in a new Decisions API at 4 cents per million input tokens and free output tokens. We intend to bring down the price even further over the coming days. Enjoy!
Grok Imagine Video 1.5 Lite is now available on fal
Grok Imagine Video 1.5 Lite is now available on falThe lightweight model in the Grok Imagine video familyText-to-video and image-to-video with native audio, 1 to 15 seconds480p, 720p and 1080p output across 7 aspect ratios, from 16:9 to 9:16
Introducing Speech (beta) By Jack Brody, Chief Product Officer·Oct 1, 2026 A new way to express yourself creatively Product Update
As new technologies make creation more accessible, more people can experience the joy of turning an idea into music, a story, or art, and get fulfillment from the simple act of making something. We call that creative entertainment, and we think it will define the next wave of consumer technology.Music will always be at the heart of Suno and what we build. At the same time, our vision has always extended to other forms of human expression. Today, we’re expanding what's possible in Suno with Speec
@bfl_ai Try it here today!
@bfl_ai Try it here today!Text to Imagehttps://fal.ai/models/blackforestlabs/flux-3/text-to-imageImage Editinghttps://fal.ai/models/blackforestlabs/flux-3/edit-image
D1 from @liquidai is live on OpenRouter.
D1 from @liquidai is live on OpenRouter.It's a competitive decision model: send your app's state and yes/no, choice, or score questions, and get back typed answers with a probability for every option. Zero data retention.$0.04/M input, $0 output, 65K contexthttps://openrouter.ai/liquid/d1
Nearly half of test subjects mistook Tavus' AI video avatar for a real person on a one-minute call
Tavus has introduced Griffin, what the company calls the first "Human Interaction Model" (HIM), a class of model designed to understand and carry on face-to-face conversations in real time. Griffin processes speech, facial expressions, tone of voice, gestures, and pauses while both receiving and generating video. In a Tavus study, 48 percent of participants believed Griffin was a real person after a one-minute video call. Previous systems maxed out at two percent. In what Tavus describes as an i
Lakebase Postgres branch-based restores for fast recovery at scale
In managed OLTP, restores have always been painfully slow and they get slower at scale. This often means that large production databases, where downtime costs the most, are left waiting the longest to recover.The usual workarounds are difficult, expensive and risky. They involve extra replicas, extra environments and even a DBA on the restore which doesn’t fully guarantee you will be protected. Failover to a healthy replica helps when a machine dies, but it does not help when the bad write is al
Your primary Bot will spot work it can take off your plate and offer to handle it.
Your primary Bot will spot work it can take off your plate and offer to handle it.It's rolling out over the next few hours, and suggestions don't count against your usage.Download Grok Bot: https://x.ai/bot
Shopify debuts Canvas, a way to build online stores by chatting with AI
Shopify introduced a new site-building tool Thursday called Canvas that lets merchants set up their Shopify store by chatting with AI. While Shopify already offered a fairly capable no-code builder before Canvas, it involved editing a selected theme by rearranging modular sections and blocks. Deeper customizations, however, would require editing code or bringing in a developer.Now, with Canvas, merchants simply chat with Shopify’s AI agent, Sidekick, to create their site. As the AI makes changes
We’re rolling out inline charts, diagrams, visualizations in Computer. Including contextual financial charts from @tradingview
We’re rolling out inline charts, diagrams, visualizations in Computer. Including contextual financial charts from @tradingview
12 stories
Introducing Olmo-core 3: Open, scalable training infrastructure for large MoEs
📄 Tech Report | 💻 Code | 🧩 Interactive demo Today we’re releasing Olmo-core 3, a significant upgrade to our framework for developing large language models featuring a redesigned open mixture-of-experts (MoE) training system. Olmo-core 3 is designed to scale MoE training into the trillion-parameter range while preserving computational efficiency. It’s one of the core systems behind the next generation of Olmo, and part of our ongoing commitment to open up the tools and training infrastructure
Hearing tech startup Legato launches its AI hearing glasses
Hearing tech startup Legato announced on Thursday that its flagship AI-based hearing glasses are now available for purchase. The glasses, called Legato Frames, integrate the company’s patented hearing-assistance technology into the arms of eyewear frames. Starting at $999, Legato Frames are designed for adults with up to moderate levels of hearing loss. The glasses stem from the startup’s goal of making hearing care more accessible by addressing the cost, comfort, and stigma associated with trad
Connecting customer context to measurable ROI with agentic marketing
Marketing leaders today face greater complexity than ever before. They work with more customer data, more channels, and more measurement tools than at any point in the discipline's history, while customer journeys have become less linear, with identity signals quickly decaying as people change devices and contact details. Marketers have spent the last two decades building commercial martech tools to keep pace - 15,000+ by martech expert estimates, not including the countless MCP integrations for
Modal Clusters are generally available
Modal 宣布 Modal Clusters 正式可用，通过一个装饰器 @modal.clustered 即可获得多节点集群，节点间经 InfiniBand verbs 通信可达 6.4 Tbps，自动配置 PyTorch 和 NCCL。
Elon Musk’s Grokipedia has a ‘newly refreshed’ design
Jay Petersis a senior reporter covering technology, gaming, and more. He joined The Verge in 2019 after nearly two years at Techmeme.Grokipedia, SpaceXAI’s AI-powered competitor to Wikipedia, recently started incorporating edits again, and today, it got some design tweaks as part of a v0.3 update, including a new logo and refreshes to its homepage and live edits page. SpaceXAI head of design Benji Taylor calls it a “newly refreshed Grokipedia.”Grokipedia’s old homepage was pretty much just a log
Sidecars: A low-latency trust boundary for Sandboxes
At Modal, our customers rely on Sandboxes to execute untrusted code written by their downstream users or, almost exclusively now, by agents. Running untrusted code isn’t a new problem: every cloud provider has to do this from day 1 to isolate their platform from their user and their users from each other. Fortunately, technologies like gVisor and Firecracker “solved” “isolation” nearly eight years ago. Unfortunately for us, they solved it for an now-outdated unit of trust. How do you protect use
VM Sandboxes: Full computers for agents
Today we’re making VM Sandboxes generally available on Modal, built for those who need to give their agents the power of a full computer.With one flag, you’ll get a fully capable Linux VM with all the niceties that you expect from a traditional modal.Sandbox, and it Just Works™. This brings the same APIs, modal.Images, sub-second cold-starts, and CPU/memory bursting capabilities as previously, all whilst supporting the hundreds of thousands of concurrent Sandboxes that our users are accustomed t
Runtime Roundup: VM Sandboxes, Multi-node clusters, and more
Modal just hosted our inaugural conference, Runtime. Here are a few of the highlights that we announced.VM SandboxesVM Sandboxes give your agent access to a full Linux computer. Agents increasingly want to live inside something that looks like a real machine: running Docker stacks, local databases and dev servers, graphical environments and mobile simulators, and even monkeying around with the Linux Kernel itself.VM Sandboxes are already in use at customers like Linear, Legora, and Snorkel power
Introducing ai_decide: make fast decisions on your governed data
Today, we’re launching a new Databricks AI Function ai_decide that makes fast decisions over your governed data.We see a lot of teams use LLMs for tasks that do not require complex reasoning and text generation. Which category does this support ticket belong to? Does this document need human review? Which model should handle this prompt?These questions power many of our customer’s high-scale enterprise workflows from processing millions of documents to controlling real-time app logic. However, u
DoorDash launches an AI agent you can text to order food
DoorDash announced on Wednesday that it’s launching a text-to-order AI agent that lets users place orders through Apple Messages. The new tool allows users to send prompts like “order my usual,” and the agent will understand that they mean their Friday night order. DoorDash says users can also ask for a specific dish and request a local recommendation. The agent will then search local spots and suggest a cart based on the prompt, and even text photos of the food it recommends. When ordering for
Customize Claude Code with mods
Today we're introducing mods, small TypeScript functions that change how Claude Code works. A mod can rewrite a prompt, add new UI, replace a built-in feature, or add entirely new functionality. You can write a mod yourself, or ask Claude Code to write one for you. Mods ship inside plugins, so you install and share them like any plugin. They work in the Claude Code CLI and desktop app.Mods run with the same access to your machine as Claude Code itself. They aren’t sandboxed, and you should only
Build Applications on NVIDIA BlueField Faster with NVIDIA DOCA Agent Skills
AI agents are becoming a standard part of development workflows, but general-purpose agents weren’t built with specialized infrastructure software such as NVIDIA DOCA in mind. Without domain-specific knowledge, agents may fall back on guesswork. This is an issue in infrastructure development because every correction cycle takes time away from deployment. DOCA is the unified software platform that unlocks the full potential of NVIDIA BlueField data processing units (DPUs) for agentic AI infrastru
19 stories
Instinct’s new product recommendations are giving some users the ick
The buzzy AI startup Instinct has just revealed a new potential revenue stream: product recommendations. Unfortunately for the company, the rollout is already rubbing some users the wrong way. Overnight, the AI agent began pushing product suggestions to users as part of a new feature called Instinct Selections, an effort to curate personalized lists for users in areas like dining, travel, and shopping. Founder Noah Shinn announced the feature’s debut later Tuesday, explaining that the idea is to
Introducing SynthID Bio
September 30, 2026 SciencePushmeet Kohli, David Stutz, Ali Cowen-Rivers and Jeremy RatcliffProof of concept for watermarking AI-generated proteins while preserving biological function.Today, we’re introducing SynthID Bio to bring watermarking technology to synthetic biology. SynthID Bio embeds an imperceptible signature directly into the biological code, ensuring the watermark is verifiable not just on a digital model but on the synthesized, physical protein itself – all while preserving its bio
Exciting update: you can now test Claude Sonnet 5.5 by @AnthropicAI directly in Arena for a limited time! Head to Direct Mode, and select...
Arena 宣布在 Direct Mode 限时开放 Anthropic 的 Claude Sonnet 5.5（High），截止 10 月 2 日上午 8 点（太平洋时间），之后仍可在 Battle 和 Agent Mode 使用。引用内容称 Claude Sonnet 5.5 是 Claude 5.5 系列第二款模型，比 Sonnet 5 快 30% 以上，多数工作成本最高降低 30%。
The ad is done. The versions aren’t.
The ad is done. The versions aren’t.Different sizes. Different markets. Same campaign.Take your ad further with Ad Variants in Luma.
Try it Ad variants in Luma - https://app.lumalabs.ai/apps/ad-variants?utm_source=x&utm_medium=p-social&utm_campaign=q3-social&utm_content...
Try it Ad variants in Luma - https://app.lumalabs.ai/apps/ad-variants?utm_source=x&utm_medium=p-social&utm_campaign=q3-social&utm_content=advariants&utm_aud=plg
Your agent video workflow can now earn with @PixVerse
Your agent video workflow can now earn with @PixVerseAn ad, a short film, a brand video: show what the PixVerse Plugin finished.Approved posts earn $10.Views can take it to $1,000 on X and YouTube, or $500 on Instagram.Join here ↓
Airbnb adds AI search, more social features
Airbnb is evolving in two directions: On the ground, it wants to compete with hotels by providing myriad services; online, it is using AI and people in its Airbnb network to aid travel discovery. As part of its fall product update, the company is adding services to the app it had launched last year, such as meal delivery, laundry, baby-gear rental, ski and boat rental in limited locations, and expanded grocery delivery. The second big feature is AI search. The company has taken a cautious approa
Perplexity Introduces Photon: A Rust-Based Retrieval Engine That Cuts p99 Latency From 800 ms to 65 ms
Perplexity has released Photon, an in-house retrieval and ranking engine written in Rust. It replaces an open-source engine Perplexity had forked for its AI-native search stack. Photon now handles retrieval and ranking for all production traffic. It also powers a new Fast Search mode in the Perplexity Search API. Perplexity reports single-call latency of 160 ms at p50 and 230 ms at p95. Is it deployable? Yes, as a hosted API. Set search_type: "fast" on POST /search and pay $1 per 1,000 requests.
One Bad Prompt Took Down a Company’s Salesforce: RSA’s Jim Taylor on Agent ID and Taming the 4,000 Shadow AI Agents Hiding in Your Enterprise
AI agents are moving into production faster than security teams can track them. They hold credentials, carry entitlements, and act on systems of record, yet most enterprises cannot say which agents are running, who owns them, or whether anyone can stop them. Gartner expects a typical Global Fortune 500 enterprise to run roughly 150,000 AI agents by 2028, up from fewer than 15 in 2025, while only 13% of organizations believe they have the right agent governance in place. At The AI Conference in S
Automate the work that keeps coming back
Go backBy Factory Team - September 30, 2026 - 3 minute readProductShareA lot of engineering work these days can feel reactive or repetitive: following up on pull requests, investigating CI failures, running security checks, or updating documentation as features ship. Other recurring tasks also take up part of everyday work: mornings may start with Slack and email catch-up; meetings need recaps, follow-up tickets, and next steps. Each task takes time, and switching between them interrupts focused
Elon Musk’s AI-powered Grokipedia is updating again
Jay Peters is a senior reporter covering technology, gaming, and more. He joined The Verge in 2019 after nearly two years at Techmeme.Grokipedia, the AI-powered online encyclopedia from SpaceXAI, appears to be updating articles once again after a months-long pause. In August, Lawfare reported that articles on Grokipedia hadn’t reviewed edits since April, but the platform’s live updates site is now showing various recent changes to pages — though as I write this, many are just a note that says “r
Agent can now work with you across even more of the creative process.
Agent can now work with you across even more of the creative process.Talk through ideas with the built-in voice agent. Sequence and edit video. Work in Blender. Queue and batch generations. Run multiple workflows at once.One creative partner across your workflow.
If you have notes, Runway Agent has it handled. With Agent Tagging.
If you have notes, Runway Agent has it handled. With Agent Tagging.Now you can simply tag Runway Agent anywhere your assets are to ask for edits, options and fixes. It's that easy.Try today at https://runway.com
Introducing dots
**Dots are remarkably capable, always-on agents built to handle everything.**They’re a whole new way to work with AI—one that gets to know what matters to you, is always working on your behalf, and takes important work off your plate so you get more of your time and attention back. Dots are frontier intelligence that have your back. Powered by GPT‑6 Astra, they have their own cloud computer, learn from feedback over time, and can work towards your goals 24/7. Through our ecosystem of plugins, th
OpenAI's reveals a new ChatGPT that looks less like a chatbot and more like an operating system
OpenAI announced a wave of ChatGPT updates at DevDay, including an open plugin system, shared workspaces, automated workflows, Slack and Teams integrations, new pricing tiers, and an enterprise marketplace. OpenAI is handing outside developers the same tools it has been using internally to build ChatGPT features. The company also added shared workspaces, automated workflows, and a marketplace for enterprise customers. Taken together, the announcements push ChatGPT well beyond its chatbot roots a
AI-powered app maker Wabi pivots to a messaging experience
Wabi, the AI startup that allowed anyone to use prompts to build apps, is undergoing a slight pivot as demand for AI agents, like Meta’s Muse and Instinct, takes off. The company announced this week that it is now becoming an AI messenger of sorts — but one that’s still capable of making apps to help you get things done. The startup’s founder, Eugenia Kuyda, who previously founded the AI companion startup Replika, described Wabi 2.0 as a “personal agent that does stuff for you and builds the int
Final Cut Camera now supports variable aperture on iPhone 18 Pro
opens in new window UPDATE September 29, 2026 Media Text of this article September 29, 2026 UPDATE Final Cut Camera now supports variable aperture on iPhone 18 Pro and even more pro options with iOS 27 Final Cut Camera, the professional filming app for iPhone, receives a major update with version 2.4, adopting a new design, adding support for variable aperture on iPhone 18 Pro, and giving creators even more pro options with the release of iOS 27. The new version is available as a free software u
New updates coming to Apple Creator Studio
QUICK READ September 29, 2026 All-New Templates for New Ideas End-to-End Cinematic Video Time-Saving Features to Boost Productivity Freeform gains several highly requested features. Boards now automatically adapt to Dark Mode, and folders let users organize boards into collections and invite others to collaborate. Writing Tools with Apple Intelligence can rewrite, proofread, and summarize handwritten notes; users can also request access to collaborate on shared files, or automate adding content
Expanding AI Storage Access with NVIDIA cuObject and the NVIDIA SCADA Server SDK
AI infrastructure engineers, storage developers, and cloud service providers need fast and secure access to high-capacity file and object storage to support AI workloads. AI workloads increasingly require high-speed data access for training, fine-tuning, inference context, tool calls, searches, and database lookups. Much of this data lies in files and objects stored both on-premises and in the cloud. Compute accelerators—including GPUs, TPUs, and XPUs—need remote direct memory access (RDMA) that
9 stories
With Dazzle, Marissa Mayer bets your camera roll has more info on your life than your inbox
When former Yahoo CEO Marissa Mayer told me earlier this month that she was finally ready to unveil Dazzle, the personal AI assistant that raised an $8 million seed round last December, I couldn’t help but wonder if she’s playing copycat to Meta’s Muse, Instinct, and the wave of similar tools that have flooded the market over the past month. But when Mayer finally gave me a demo, Dazzle proved it’s taking a different approach. Instead of building context about you from text-heavy apps like email
Manus 2.0 lets users edit videos, host multiplayer games, and run agents remotely from their phone
Manus is turning its AI agent into a platform with version 2.0, letting users edit videos, host multiplayer games, and run personal agents with their own phone numbers. In one tested configuration, the new Cascade agent harness used 23.2 percent fewer tokens and cost 32 percent less to run than the previous system. Automations can now fire on events like incoming emails or Slack messages. With Computer Use, Manus works with approved files and apps on the user's own machine, which users can also
September 28, 2026•Customer Stories Bonjour Turns One Script Into Every Ad Style with Runway
Specializing in mushroom- and plant-based adaptogenic beverages, France-based Bonjour has built its growth on paid social. Keeping ads fresh enough to compete with the organic feed is the responsibility of a 15-person creative pod. They build every new concept out in several animated styles and test each one against real ad spend before deciding what actually works. With Runway handling that animation, the loop from script to verdict runs in four to five days, work that once meant hiring a freel
Lakebase Search: State-of-the-art full text and vector search for Postgres
Traditional OLTP systems weren't built for the search demands of AI agents. They require low-latency, high-accuracy retrieval across all your data and often execute massive parallel searches. Until now, solving this meant duct-taping a standalone search engine to your primary database with an ETL pipeline.But what if your OLTP database could just run the search workload efficiently?Today, we are bringing a fast and scalable search engine to Lakebase Postgres via two extensions: lakebase_vector (
Shopify opens checkout to browser-based AI agents
While some retailers, like Amazon (and Adidas, apparently!), are blocking AI agents from making purchases on users’ behalf on their respective platforms, e-commerce platform Shopify has moved in the other direction. On Monday, the company announced that browser-based AI agents can now complete purchases on Shopify merchants’ sites, extending their capabilities beyond just searching for products and adding items to carts. Shopify previously supported WebMCP for its storefronts and carts, allowing
NVIDIA Launches Open Agent Safety Platform: OpenShell Sandboxes Agents on Vera CPUs While Sentry on BlueField-4 Quarantines Them in Milliseconds
NVIDIA has launched the NVIDIA Open Agent Safety Platform, an open software platform and reference system design for AI agent security. It pairs the OpenShell secure runtime with NVIDIA Sentry, an out-of-band watchdog on BlueField-4 DPUs. The core idea is simple. Safety controls should not live inside the agent they are meant to control. Today, with over 100 industry partners, we introduced the NVIDIA Open Agent Safety Platform, bringing together OpenShell and Sentry.Artificial intelligence is e
MicroLLM Lab – Try 7 tiny LLM's in the browser
max new Next action: Objective checks (regex / exact tokens), not writing quality. A 135M model is allowed to fail — that is the measurement. Pick models, then run. Estimate uses your last tok/s if we have one. Next action: Speed (tokens/s, sustained decode, suite wall) and accuracy (pass rate on objective tests) from runs in this browser. Numbers stay on this machine. Charts use the latest suite per model. Your Name or Handle Device & Hardware Information Next action: Write a benchmark in JavaS
Nvidia launches new platform for reining in rogue AI agents
As the debate rages over whether the recent spate of rogue AI agents is a step toward AGI or a more conventional engineering problem, Nvidia is offering its own answer to the problem. Nvidia CEO Jensen Huang on Monday introduced a toolkit of software and hardware products that add independent security layers around AI agents to ensure they stay within their test environments even if they attempt to break out. The release follows a string of hacking incidents involving AI models from Anthropic, G
AA-AgentPerf-Local: Benchmarking local AI agents on laptops and workstations
Announcing AA-AgentPerf-Local, our open-source inference testing tool for local AI models - test how fast agentic AI can run on your own laptop or workstation, and browse our list of serving configurations to plan your next agent setup Key points: ➤ We’re open sourcing AA-AgentPerf-Local, which replays real agent trajectories on laptop & workstation hardware to test inference performance ➤ We’re releasing initial results for NVIDIA DGX Spark, NVIDIA GeForce RTX 5090, AMD Ryzen AI Halo, and MacBo
9 stories
Nvidia wants to keep AI agents on a short leash with a watchdog built into its chips
OpenAI isn't the only lab dealing with this. Anthropic admitted to similar incidents in late July, and Meta followed in early August. It also recently came out that Google's Gemini hacked three real companies during a test back in May. OpenAI, Anthropic, and outside researchers are now reviewing tens of thousands of other cases. According to OpenAI, many of them are just routine research activity. Nvidia's technology isn't entirely new. The platform combines OpenShell, open-source software Nvidi
Nvidia says its new AI safety platform can contain rogue agents within ‘milliseconds’
Skip to main contentThe Open Agent Safety Platform is designed to enforce AI boundaries.The Open Agent Safety Platform is designed to enforce AI boundaries.by Emma RothSep 28, 2026, 1:36 PM UTCImage: Cath Virginia / The VergeEmma Roth is a news writer who covers the streaming wars, consumer tech, crypto, social media, and much more. Previously, she was a writer and editor at MUO.Nvidia is launching a new safety platform designed to contain and monitor AI agents, a move that comes in response to
NVIDIA Launches Open Agent Safety Platform to Secure Agents From Testing to Deployment
Open Software Platform and Reference System Design Brings Together Industry, Researchers and Public-Sector Organizations to Set Safer Boundaries for AI Agents, Share Best Practices and Foster International Cooperation to Raise the Bar for Safer AI Agent Deployment News Summary: NVIDIA Open Agent Safety Platform consists of NVIDIA OpenShell open source software and the NVIDIA Sentry reference system design that enables full-stack governance and control across software and the hardware, compute an
20 Agentic Use Cases of TypeSafe AI’s Jev
Last week, TypeSafe AI released Jev, its first System One model. Founder Diogo Almeida previously worked at OpenAI on the instruction-following research behind ChatGPT. Jev does not chat, write code or summarize. It takes unstructured state and returns typed decisions with calibrated probabilities. That makes it a natural fit for the thousands of small judgments inside an agent loop: which model to call, whether a command is safe, which passage is relevant, whether the agent is actually done.How
Product·Sep 28, 2026 Team Bots: shared AI teammates that learn as they work
Today we’re launching Team Bots, Grok Bots that work and learn alongside your team. Give one access to the files, apps, and expertise it needs, then share it so everyone can work from the same context. At SpaceXAI, Team Bots brief account teams each morning, coordinate engineering projects, and answer data questions across the company. Here’s how they work, how we use them, and how to build one for your own workflow. Team Bots bring context, tools, and memory together You build a Team Bot around
Manage and de-risk your API keys with new Security Center
Dealing with a leaked key is stressful. The fewer old keys you have, the fewer can leak. That’s why we built Security Center. It’s one place to see every key across your workspaces, spot the risky ones, and disable, archive, or cap hundreds at once. It’s available on all OpenRouter plan types. Try it under Settings > Security. 1,000+ keys across 85 employees The obvious defense against leaked keys is to have fewer of them. So we audited our own OpenRouter org, and because we build OpenRouter on
Inside Fireworks Multi-region Deployments: One Deployment, Global Scale
TLDR; Discover when to leverage the GLOBAL option and how we architected multi-region routing to handle global workload placement off the inference hot path, optimizing for low latency, resilience and operational ease.No single region is an infinite GPU pool.Fireworks has a full spectrum of serving options, and operates on a global scale across 30+ regions across dozens of cloud providers in one of the industry’s largest independent GPU fleets. Yet, a massive global footprint only protects your
Introducing FireRouter with Opus
Fireworks Nexus enables engineering teams to drop leading open models in the harnesses they already use and cut spend in half without sacrificing speed or quality. The solution includes FireRouter, the first cache-aware router on the market, which makes a big difference in speed and cost. Today, we’re introducing FireRouter with Opus, optimized for the Opus family and now available in both our CLI and, for the first time, as a standalone router model. Any Fireworks account can point to it as a s
NVIDIA Open Agent Safety Platform: A Reference for Continuous In-Silicon Agent Monitoring
To understand where agentic AI stands today, consider the last seismic shift in technology: the rise of the internet in the 90s. It was new and full of possibilities. You could build a website over a weekend and share it with the world, or chat with someone half way around the world in online chat rooms without long-distance telephone fees. It brought endless opportunity, but also a lot of risk. A website could run code on your machine, steal your sensitive information, or infect your computer w
1 story
3 stories
Introducing the new Copilot with Home, Code and Autopilot
Microsoft 官方博客宣布推出新版 Copilot，新增三项能力：Home 整合 Chat 与 Cowork 并内置 Word、Excel、PowerPoint；Code 让非开发者用自然语言构建应用，与 GitHub Copilot 共用底层技术并在沙箱内运行；Autopilot（原 Scout）是云端常驻的主动式智能体。
The Biggest News From Connect 2026
Meta 在 Connect 2026 宣布将个人 AI 智能体 Muse 带入 AI 眼镜，并新增 Walmart、Sephora、GitHub、Instacart 等更多 connectors，Muse 还将拥有自己的邮箱地址。
Build plugins for Claude
Anthropic 推出 Claude 插件（Plugins），作为第三方开发者为 Claude 构建扩展的主要方式，插件可打包 MCP 连接器和 Agent Skills。
1 story
1 story
3 stories
The new Mac mini and Mac Studio are available today
Apple 宣布新款 Mac mini 和 Mac Studio 于 9 月 22 日开售。Mac mini 搭载 M6 和 M5 Pro，AI 性能最高提升 4 倍。
Announcing vllm-metal: Concurrent Serving on Apple Silicon
vLLM 官方发布 vllm-metal v0.28.0，把 vLLM 的 V1 调度器、paged KV cache 和 OpenAI 兼容服务器带到 Apple Silicon，由 MLX 和 Metal 执行模型，版本号与上游 vLLM 对齐。
Transformers now runs llama.cpp quants
Hugging Face 宣布 transformers 支持直接运行 GGUF 量化模型，通过 from_pretrained 传入 gguf_file 即可加载 Hub 上的 GGUF checkpoint，并复用 ggml 的 Metal 内核。
1 story
2 stories
You can now train and run 500+ models locally with our Unsloth Docker image! 🐳 Use our new GUI or ...
Unsloth 宣布可使用其 Docker 镜像本地训练和运行 500+ 模型，提供新 GUI 和 notebooks 工作流，无需配置，支持 NVIDIA 和 AMD，指南见 https://unsloth.ai/docs/get-started/install/docker。
Introducing the Life Sciences Verification Program
Anthropic 推出 Life Sciences Verification Program（LSVP）beta，向经过验证的生命科学专业团队开放 Mythos、Opus 和 Sonnet 模型上比通用版更宽松的生物相关访问。
Showing the latest 81 of 759 stories.