XIO

xWire

Posts from every feed QUEST has found across the web — blogs, news, podcasts, channels and more.

AI Infrastructure Archives - The New Stack

Respan spent two years building on LiteLLM. One upgrade made it start over.

Respan released an open-source AI gateway written in Rust on Tuesday, after spending two years building more than 100,000 lines of wrapper code around LiteLLM. The new gateway replaces that system with a small core and plugins that handle model providers, routing, MCP tools, realtime voice, and security controls. A LiteLLM upgrade breaks The breaking point came in January 2026, when a routine LiteLLM upgrade caused intermittent connection failures. Respan rolled back the update but never…

Assess the entire inference stack — not just open model benchmarks

Impressive benchmark results should make it easy for engineers to select the right open model for the outcome they are tasked with developing. However, that is not the case. Benchmarks indicate an open model’s optimal performance, but in production, a model must work repeatedly. The challenge is that in production, other factors matter, such as the information provided to the model and its accuracy. So benchmarks can’t tell an engineer whether the AI they are developing will complete the task…

DeepSeek open-sources the inference engine behind its API. And Huawei gets equal billing with Nvidia.

DeepSeek has open-sourced the inference engine it uses to serve its models in production, a step it said last year wasn’t yet practical. Dubbed DeepInfer, the framework is designed to let developers run DeepSeek’s V4.1 Flash model on their own infrastructure, with many of the same optimizations the company uses behind its API. Support for future DeepSeek models is also planned. “Running a model efficiently is a different matter altogether.” DeepSeek has, of course, long made its model weights…

Beam pulls single app windows off the Mac Mini running your agents

Beam is a new Mac app that takes a window running on one Mac and puts it on another, where it behaves like a locally installed app. Why? Agents. Serial founder Hiten Shah, who co-founded Crazy Egg and KISSmetrics, built it with Sam Asante under a company called Noats. It launched on Product Hunt on Tuesday and is aimed at the headless Mac Mini sitting in a closet running AI agents. The pitch is great: Instead of opening Screen Sharing and steering a full remote desktop, a user picks a single…

DeepSeek made its name with cheap AI models. Now it’s undercutting search API rivals, too.

Not content with challenging the major frontier AI labs with lower-cost models, Chinese AI company DeepSeek is setting its sights on the search API market.On Saturday, the company announced it has opened up its internally developed search engine to outside developers through two new APIs, Search and Fetch, allowing them to retrieve live web search results and webpage content for use in their own apps and agents. Both services cost $0.75 per 1,000 successful requests, and developers can use them…

Running local AI models on Microsoft’s Surface Laptop Ultra: What we found

Microsoft is positioning the new Surface Laptop Ultra as an alternative to Apple’s MacBook Pros for creatives and developers — and especially AI developers. Built on Nvidia’s N1X-based RTX Spark platform, with a Blackwell RTX GPU with up to 6,144 cores, a CPU with up to 20 cores, up to 128GB of unified memory, and a price that undercuts Apple, it’s a tempting alternative if you want to run local AI models and also want to have a powerful machine for editing videos and gaming on the go. This is…

Vercel adds always-on workers to run more of your backend

Vercel is giving developers more options for running their backends on its platform. At its Ship conference in San Francisco on Thursday, the company introduced Daemons, a new service for continuously running background workers, and made its Python support generally available. It also made Drives, its persistent storage for AI agents, generally available. Always-on workers, pull-based jobs Daemons, which is now in public beta, lets developers deploy a container and specify how many instances…

Anthropic’s been eyeing confidential AI. It just made its move with another acquihire.

Anthropic’s done it again. On Tuesday, Opaque Systems announced a new agreement with Anthropic that “carries Opaque’s work forward.” Vague wording aside, the agreement marks another acquihire for Anthropic, with the AI juggernaut licensing Opaque’s technology and bringing on board some (but not all) of its employees, including Opaque CEO Aaron Fulkerson and CTO Rishabh Poddar, reports The Information. The agreement gives Anthropic access to expertise in confidential AI, the “real-world…

Agentic AI doesn’t have a model problem. It has a retrieval problem.

On BrowseComp-Plus, GPT-4.1 answered 93.49% of the questions correctly when it was given the supporting documents directly. When an agent used the same model but relied on BM25 to search for evidence to pass to the model, accuracy fell to 14.58%. That difference matters because a system that gets the answer right 93% of the time is far more useful than one that gets it right only 15% of the time. An agent can look excellent in a controlled demo. The corpus is clean, the task is familiar, and…

Researchers left a hidden message for rogue AI agents. Then the agents wrote back.

A team of researchers investigating AI agents trying to bypass Alibaba’s anti-bot protections left a hidden message inviting the agents to contact them. The agents wrote back. The investigation began when Swarmchasers researchers Ethan Elasky, Rowan Howard-Jones, and Jonathan Elsworth Eicher helped uncover thousands of attempts by AI agents to retrieve data from Amap, Alibaba’s mapping service, which Alibaba says reaches nearly 1 billion monthly active users. The activity traced back to agents…

Your database decision could come back to haunt you

When building AI prototypes, developers naturally want to get up and running as fast as possible. But move too fast, and those decisions can come back to haunt you if they lead you to make flippant decisions about data infrastructure. For smaller, less-regulated startups building SaaS or internal applications, the consequences may not be as dire. Wherever you start building, you can usually keep running there even when applications move into production. But for large-scale enterprises operating…

Workday goes headless, opening its HR and finance data to outside AI agents

Workday is opening up its platform and allowing developers to build their own HR and finance agents in Claude Code, Cursor, and other tools, as long as those agents follow the same rules as the people using them. On Tuesday, at its Workday Rising conference in Las Vegas, the company made its Trusted Agent Platform “headless.” With this, its Model Context Protocol (MCP) server, SQL access to live Workday data, and a shared runtime for agents are now generally available, and customers can run…

Kubernetes on cgroup v1 is dead. Here’s what comes next.

Welcome back to another edition of Road to KubeCon, where we’re tracking the latest movements in the cloud-native ecosystem as we prepare for KubeCon + CloudNativeCon North America 2026, happening November 9-12 in Salt Lake City, Utah. In this edition, we consider what happens when compute workloads need to run AI on the edge. Underlying Kubernetes mechanics change, as do observability practices and security decisions. We take a look at the emerging tactics and how a couple of key projects are…

AI data lakes are driving new storage demands

The frenetic pace of AI development is challenging data center design and operational paradigms. The twin pillars of hyperscale computing – intelligent processors and the memory that feeds them – are no exception. Hyperscale operators have invested heavily in GPU clusters and high-bandwidth memory to drive AI model training and inference. While those investments have pioneered historic performance gains, they also exposed a design rift that data center architects are racing to close. One of…

Decision models are suddenly everywhere. OpenAI’s is now public.

This week, OpenAI opened its Decisions API to public beta and added image inputs to make fast decisions from visual content. First announced last week as a likely response to TypeSafe’s Jev, OpenAI is now billing its decision model at $0.10 per 1M input tokens. There are no output-token charges, nor charges for cache reads or writes. Running on GPT-6 Luna, the public beta version gives all developers the ability to turn text or images into three types of structured outputs: predicates (i.e.,…

The CNCF is graduating projects faster than ever. AI agents are helping with the due diligence.

Back in 2018, a scrappy startup called OpenAI used Kubernetes to balance compute loads across its own data centers and AWS and Azure capacity. Years before the release of ChatGPT and the start of the current artificial intelligence boom, AI companies were training the future with the help of open-source technology. Jonathan Bryce, Executive Director of the Cloud Native Computing Foundation (CNCF), which stewards Kubernetes and other software tools, tells The New Stack that Kubernetes and other…

Anthropic’s answer to Dots and Muse is already inside Claude

I’m Matt Burns, Chief Content Officer at Insight Media Group. Each week, I round up the most important AI developments, explaining what they mean for people and organizations putting this technology to work. The thesis is simple: workers who learn to use AI will define the next era of their industries, and this newsletter is here to help you be one of them. OpenAI launched Dots at DevDay on Tuesday. Each Dot is an always-on agent with its own cloud computer and browser, hooked into the…

Microsoft Fabric is where AI agents learn how the business works

Microsoft is turning Fabric, its integrated data platform, into the place enterprise agents go to learn how a company works, whether that’s what happened last quarter or what’s happening on the factory floor right now. On Tuesday, at its FabCon and SQLCon data conference in Barcelona, Spain, Microsoft outlined the next step in its vision for how the products that make up Fabric can simplify all of this information gathering. Microsoft, for example, described how Fabric IQ, the context layer…

Eclipse wants companies to be free to switch AI providers. Today, doing so can mean a costly rebuild.

If you rely too heavily on one AI provider, switching can mean rebuilding workloads and moving data while you face untangling integrations. A new vendor-neutral group aims to help organizations keep their options open. That’s the goal of the Eclipse Foundation, which announced the formation of the Sovereign AI Foundation on Wednesday. The new body will serve as a vendor-neutral initiative to help member organizations manage AI platform lock-in, monitor cyber risk, and maintain control of their…

Ember-1 vs. Kimi K3: Nearly identical results at 3.4 times the speed

Fireworks Research launched Ember-1 on September 23 as a research preview. Built on Moonshot’s open-weight model, Kimi K3, it claims it matches Kimi K3’s quality using roughly 40% fewer tokens. Fireworks says it “learned to cut unnecessary reasoning while keeping the thinking that matters.” Reasoning tokens bill as output, so a model that thinks less should cost less. There’s a catch, though. On OpenRouter, Ember costs $3 per million input tokens and $15 per million output tokens. Fireworks…

Enterprise AI desperately needs to protect data and models. Here’s how confidential AI could do it.

Most people already understand what generative AI can do. But enterprises run into problems when they need to give a model access to information that cannot leave their own environment, such as a patient record, a customer’s financial details, or a company’s most valuable intellectual property. Sending that data to a cloud or SaaS service means it crosses external networks and is processed on infrastructure run by another organization, creating additional concerns about control, accountability,…

Elon Musk says space will soon hold nearly all compute. Google is still finding out if its chips can work there.

While SpaceX and Nvidia plan to place a space-optimized Vera Rubin AI system in orbit by late 2027, Google’s far more modest plan, Project Suncatcher, is to launch a test satellite with Tensor Processing Units to see whether AI data centers in space are viable. For SpaceX and Nvidia, it’s a race to get ahead in AI data centers in space, with Starmind AI 1 satellites. Never mind that the engineering challenges of cooling, radiation resistance, and repairability still haven’t been sufficiently…

Microsoft’s new Copilot agents get their own email, calendar — and a place in the org chart

Microsoft announced what it calls its biggest Copilot update to date on Friday, with CEO Satya Nadella describing Copilot as “a new OS for work.” Nadella framed Copilot as spanning every model, form factor, and task, and the update puts Autopilot, which Nadella called a “proactive and long-running agent built for the enterprise,” at the top of his list of the update’s four components. The pitch targets office workers, but the more consequential change for developers is the infrastructure…

OpenAI and Cursor agree on agent coordinators. They disagree on who runs them.

OpenAI opened its Agents API in public beta this month, exposing the harness that powers Codex with managed sessions, tool coordination, and subagent orchestration. On the same day, September 10, Cursor launched Projects to coordinate multiple coding agents around larger bodies of software work. While the products sit at different points in the stack, both converge on the same architecture: a coordinator understands the larger objective and manages the work, while specialized agents execute…

What managing 150,000 AI agents could look like for database teams

The database administrator of the future will spend considerably less time administering databases. That sounds contradictory, but AI agents are taking over that work. For decades, DBAs have handled the decidedly hands-on work of keeping databases available, performant, secure, and affordable. They provision capacity, troubleshoot slow queries, manage migrations, and step in when something inevitably goes sideways. AI is already taking on some of that work. At the same time, it is creating a…

Query decomposition doesn’t fix context starvation — it just moves it

I have a small example that would best communicate the message I am trying to convey: say you built a chat widget for GitLab’s public documentation (the corpus we are experimenting with in this article) and one of the developers sends this kind of message: We got an email saying our card was declined for something called “quarterly reconciliation” and I need to know what actually happens now. On top of that, I think we’ve gone over our seat count; there are more people in the group than seats…

Q.ANT gives away the software for its light-powered AI chips in a CUDA-style bet on developers

Q.ANT, a startup out of Stuttgart, Germany, builds processors that use light instead of electricity to do some of the math behind AI. The company pitches them as a way to run AI on a fraction of the power today’s chips need. Now developers can start writing software for those chips without owning one. Q.ANT pushed a free, open-source software kit to GitHub this week that lets developers build and test programs on a normal computer, then run them on the real chips once they get access. This is a…

A third option is emerging in the fight over AI and your data

Not your keys, not your coins. Not your model, not your data? Over the summer, the tech industry was consumed by a debate about AI use in the enterprise and the need to protect IP. If an enterprise used proprietary models, was data leakage a necessary evil? Companies seemed to have two options: They could use state-of-the-art, proprietary models and risk losing control of their data, or they could use open-weight models and never kiss the frontier. Thankfully, a third option is emerging.…

Claude Opus 5.5 wants to finish your coding tasks, not just start them

Anthropic wants developers to use Claude and its family of tools to handle complete coding tasks. The company introduced Claude Opus 5.5 on Tuesday to span more of the software application development lifecycle: from design specification creation through debugging to code generation and testing. As the first release in a new family of Claude 5.5 models, Anthropic says that Claude Opus 5.5 performs at the level of Claude Fable 5.1 on “most work” tasks and costs around 40% less to run than Opus…

Your AI agent is burning tokens on choices that don’t need words

AI agents spend a ridiculous amount of compute generating text nobody actually needs. The decisions an agent makes along the way don’t require a written answer and yet, agents still send them to generative models, wait for an answer while burning through tokens and then parse that output back. The overhead is already drawing scrutiny — OpenAI’s own researchers recently disclosed spending $7,000 a day running agent workloads.Kev, a new family of open decision models built on Qwen 3.5, takes a…