What the 45 sources on blogs to watch have published, grouped by subject rather than by publisher, and rebuilt every hour.
Updated 41 minutes ago, from 45 sources, holding 1500 posts.
The page shows the last 90 days, which is 1500 posts of the 1500 the index holds, across 18 topics.
Everything in the window, newest first, whatever topic it was filed under.
Learn how LangChain built internal agents across marketing, sales, and engineering, and how they have completely changed how those teams work.
#20
Qualcomm’s Snapdragon X2 Elite uses a hybrid core setup, breaking away from the all-big-core strategy that the company previously used with in-house core designs.
We have Fable at home... or so we are being told by paid shills
Release: battledrones v14: Compass Rose, a four player map I added four player online mode to Battledrones - previous maps were limited to two. Here's the initial prompt I used: Add a new map for a four player battle - four way…
You can now view and manage your team's Spend Management budget from the Vercel CLI. Run vercel budgets and you, or the scripts and agents you run, get back your current spend and the active budget. View, set, and remove a budget This…
Tool: Quick Study Prompt to Claude Opus 5.5 High : Build an artifact where you open a photo and it rapidly (in a few seconds) paints a version of that photo using a canvas element and simulated paintbrush See also my previous experiment…
Vercel MCP now includes submit_feedback , a tool for sending bug reports, friction reports, and feature requests directly to the Vercel team from your agent or harness. Ask your agent to give feedback in plain language: Send Vercel…
An agent working on production systems needs permission to act. A request like “fix this deployment” doesn’t tell you upfront which actions it will take or what access it will need. Grant agents broad access and a mistake can affect…
Introducing Kimi K3.1 Pro Kimi K3.1 Pro landed on the Kimi API today - it's not yet open weights, but Moonshot promise "Full model weights will be released by Nov 4, 2026". They're also offering a "Highspeed" version that's 2-3x faster,…
Deployment summaries now show the size of the build cache each deployment saved and how close it is to the limit for its build machine . The same information is in the CLI with vercel inspect on version 65.1.0 or newer, and in the…
AI agents now write a growing share of production data pipelines (over 70% of SDP...
By Roberto Perez Alcolea , Java Platform Team Quasar connects a proposed API change with observed references in consumer repositories, giving the producer evidence for a release decision At Netflix, shared Java libraries support…
Naive byte-range caching can waste gigabytes of origin bandwidth on small requests. We rebuilt our cache architecture in Pingora to coordinate partial fetches automatically, cutting R2 p99 latency in half.
Cloudflare's new post-quantum GeoKey Manager protects geographically fenced decryption keys from harvest now, decrypt later attacks. Learn how we redesigned our cryptography stack to support strict data sovereignty in a post-quantum world.
Multi-Turn Reinforcement Learning (MTRL) on Amazon SageMaker AI trains models on the outcome of a whole multi-step task rather than labeled steps. This hands-on guide walks the full pipeline using Wordle: deploying the agent and reward…
Access a network of highly vetted partners to build custom RL environments to accelerate your post-training efforts.
The Global Performance Tuner is a feature for automatic performance tuning released in NVIDIA TensorRT 11.1. It provides automatic, accuracy-aware and...
Deploying a trained neural network involves more than translating each operator into a GPU kernel. It is a compilation problem with many legitimate answers. For...
Agents waste time and context on clumsy searches, oversized results and stale data. Learn the retrieval techniques LangChain's Deep Agents harness uses to get the right context in.
Vercel Connect can now create a separate restricted Stripe API key for each agent or task. When you create a Stripe connector, you can now choose Managed API Key as the connection method. Vercel creates the key in your Stripe account,…
Learn how OpenAI scaled Habitat with reliable multi-tenancy, global caching, and Azure Cosmos DB to deliver 12× faster reads for ChatGPT.
Updating a SageMaker AI endpoint that has Application Auto Scaling registered makes AWS CloudFormation and AWS CDK deployments fail. This post shows how to build a reusable AWS CDK construct with CloudFormation custom resources that…
Amazon Quick Sight layer maps now support Point and LineString geometry from the same GeoJSON workflow, so you can plot store locations and the routes between them. This post covers the dataset and GeoJSON structure layer maps require,…
Bayer Crop Science built the Agentic Decision Engine (ADE), a multi-agent platform on Amazon Bedrock AgentCore that lets breeders, analysts, and researchers ask natural language questions about their analytics platform. This post covers…
AI models are solving more mathematical problems. A pattern is emerging in their successes that could help us predict which problems they’ll solve next.
Teaching a robot to navigate an obstacle course requires more than a walking policy. It must perceive its surroundings, maintain balance, and adjust its actions...
Learn how Cerebras serves OpenAI’s GPT-5.6 Sol at up to 750 output tokens per second—without shrinking the model or compromising intelligence.
Shipsidekick moved from Neon to ClickHouse Managed Postgres in a single weekend, cutting CPU usage from 100% to 5-10% and making queries about 10x faster.
the Cyber onslaught continues.
OpenAI explains how we strengthened containment for frontier model training and evaluation, including offline environments and high-compute red-teaming.
Interfaze on Vercel 50 million requests a month Backend running on Fluid compute Multiple model providers behind one fallback layer with AI Gateway Interfaze builds models for deterministic work that general-purpose LLMs handle poorly:…
The Vercel CLI now includes vercel feedback , which sends bugs, friction reports, and feature requests to the Vercel team from your terminal. Coding agents can also send feedback when they run into issues. Use vercel feedback with -f or…
Day-0 vLLM support for MiniMax-M3.1-Flash on NVIDIA Blackwell, NVIDIA Rubin, and AMD MI350-series GPUs, with NVFP4 MSA kernels, NVFP4 MegaMoE, DSpark speculative decoding, AgentX results, and day-0 RL.
OpenAI and Poland’s NASK are partnering on AI safety, security, and technical research to strengthen safeguards and support a more resilient digital ecosystem.
In Databricks’ State of AI Agents 2026 report, there’s a chart from Neon’s telemetry that’s somehow both mind-blowing and unsurprising. In October 2023, AI agents created 0.1% of the databases on Neon. By October 2025 they created 80%…
In this blog you will learn how to port an FP16 GEMM kernel to FP8 on AMD RDNA™ 4 GPUs. You will see what the FP8 WMMA instruction changes, how the smaller element size lets larger tiles and deeper K slices fit in LDS, and how four tile…
Kimi K3.1 Pro from Moonshot AI is now available on AI Gateway , along with Kimi K3.1 Pro Highspeed , a faster version for interactive use. K3.1 Pro is the first model in Moonshot's K3.1 series. It builds on Kimi K3 with improvements in…
vLLM serves GLM-5.3 on 16 H200s with PCP and DCP prefill and Hybrid HiSparse decode, reaching 1M-token contexts and 1.7–3.2× the per-user speed of data-parallel P/D on the AgentX agentic benchmark.
gsplat is a GPU-optimized Python library for training and rendering 3D Gaussian Splatting (3DGS) models. In earlier posts we showed how to train gsplat scenes on AMD GPUs, took a closer look at the mathematics of 3DGS, and applied…
With P/D disaggregation and Wide-EP on a GB300 NVL72 rack, vLLM serves DeepSeek-V4.1-Flash on SemiAnalysis AgentX at 954K tokens/s per GPU at a P90 of 196 tokens/s per user, 7.9x our aggregated single-node result.
Every Pro team can claim a free domain for one year. That offer now includes .co , .studio , .company , .info , .pro , .live , .digital , and .agency , bringing the eligible TLDs to sixteen. The offer still covers one domain per team,…
Samartha Chandrashekar & Nick Parker You click play on a Friday night, and in the second that follows, dozens of services answer in milliseconds. Most of them run on Titus, our internal container platform. Titus is, in essence, a…
Every organization has unique requirements for building and scaling generative AI. At Google Cloud, we are committed to giving developers model choice and architectural flexibility to run workloads however they need. Today, we’re…
DE STAAT - FREEDOM (Official Music Video) OK well this is a first for me... at 00:44 in this new music video by Dutch alternative rock band De Staat : The visual effects are credited to Matties Grooten of Fake it or Leave it . Via flir…
Fast, focused, low-cost decisions for any application.
BackgroundAccording to the Communications Fraud Control Association's 2025 survey...
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)…
Learn how Cerebras maps trillion-parameter frontier models across WSE-3T wafers to deliver scalable inference at 1,000–2,000+ tokens per second.
The Amazon Polly Generative engine gets a major upgrade: more expressive, more accurate speech with low latency, plus 14 new voices across 9 locales and general availability of Hindi and Finnish. This drop-in upgrade requires no code…
Learn how IskayPet, a leading pet care company in Iberia, built a multilingual corporate knowledge assistant on Amazon Bedrock Managed Knowledge Bases. The assistant answers employee questions across retail, veterinary, onboarding, and…
Amazon SageMaker AI serverless model customization now includes training sessions, generally available today. You write and coordinate your own post-training loop (custom loss functions, reward signals, or multi-turn reinforcement…
Amazon SageMaker AI training session APIs let you write your own LLM training loop (SFT, RL, or custom objectives) in a Python script while the service runs the GPUs. This post introduces the programming model and three SDK clients,…
Amazon Quick is a HIPAA-eligible service, but eligibility is not the same as compliance. This post explains who HIPAA applies to, how PHI flows through Amazon Quick, what the platform provides for regulated environments, and a…
Voice agents often wait for an agent's full response before starting speech synthesis, adding latency users feel directly. This post shows how to stream each sentence to a Kokoro-82M text-to-speech model on Amazon SageMaker AI as the…
Elastic GPUs, sandbox growth, and the economics of the agent cloud
Elastic GPUs, sandbox growth, and the economics of the agent cloud
As Cybersecurity Awareness Month continues, the GitHub Bug Bounty team spotlights @r31n, diving into their research approach, favorite bug classes, and advice for aspiring hackers. The post How bug bounty sharpens one researcher’s…
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…
GPU applications are processing ever-growing datasets with increasingly powerful accelerators. According to the 2025 Stanford AI Index, compute for notable AI...
Teams deploying custom text-to-speech models on Amazon SageMaker AI often guess at autoscaling thresholds, and a wrong guess either wastes GPU spend or breaches a latency SLO. This post deploys Kokoro-82M as a custom container and…
Rescale used the AI-Driven Development Lifecycle (AI-DLC) framework on AWS, with Strands Agents and Amazon Bedrock, to build an intelligent test-triage agent in weeks. Learn how AI-DLC structures AI-assisted development and how Rescale…
AI developers aren’t just running a single model anymore—they’re orchestrating multiple models in parallel, chaining inference calls, and experimenting...
As fun as AI is, we built so much bad software over the last 30 years, and now the universe is punishing us for our sins. — Armin Ronacher Tags: security , generative-ai , armin-ronacher , ai , ai-security-research , llms
API Gateway now supports real-time streaming—including Server-Sent Events, WebSockets, gRPC, and chunked transfer encoding—in Public Preview without requiring changes to your existing API specifications. This update allows developers to…
Samartha Chandrashekar & Nick Parker You click play on a Friday night, and in the second that follows, dozens of services answer in milliseconds. Most of them run on Titus, our internal container platform. Titus is, in essence, a…
How Moshi turns market data into actionable insights, and why AI tools have to fit the trader’s workflow.
How Moshi turns market data into actionable insights, and why AI tools have to fit the trader’s workflow.
How Moshi turns market data into actionable insights, and why AI tools have to fit the trader’s workflow.
LangSmith LLM Gateway is in public beta: spend caps, rate limits, model fallbacks and PII redaction for production agents, without provider lock-in.
Cloudflare has achieved South Korea's ISMS certification following an audit across 80 controls. Learn how we localized user consent flows, adapted social sign-up paths at the edge, and validated our global security program.
Spindle is an open-source API for training and fine-tuning open models — deploy a server with a single command and run any Tinker-compatible training code against it.
LlamaParse is the most accurate agentic OCR Platform. Parse and Extract your complex documents at the Pareto frontier of cost and accuracy.
Learn why standard OCR falls short for KYC compliance and how agentic document extraction delivers the field-level accuracy AML regulations require.
Learn how mortgage document automation transforms loan processing, from document ingestion and extraction to validation and system integration.
Learn how KYC automation replaces manual verification with scalable, compliant workflows that cut costs, reduce errors, and speed up onboarding.
Learn why single-pass extraction fails and how deep extraction uses agentic verification to deliver production-grade accuracy on complex documents.
Learn how AI document classification automates document sorting and routing at scale, and what separates systems that perform on real documents from those that don
Learn what impacts OCR accuracy, how it’s measured, and how to improve performance in real-world document processing workflows.
Learn how unstructured data extraction turns documents, PDFs, and text into structured insights using AI, NLP, and LLMs for scalable data processing.
Agentic document extraction uses AI reasoning and visual grounding to accurately process complex documents without templates. Learn how it works.
Agentic document processing uses AI agents to autonomously handle document workflows end to end. Learn how it works and where to start.
OCR for tables converts complex document layouts into structured, machine-readable data. Learn how LlamaParse preserves table integrity.
OCR for receipts breaks when layouts vary and rules pile up. Discover how agentic OCR reconstructs line items, totals, and structured data for automation.
OCR for images helps convert photos, labels, and screenshots into structured text. Compare the top AI OCR tools and learn what makes a reliable image-to-text system.
The Big Hardening begins
The Big Hardening begins
Discover how OCR for invoices streamlines finance operations, automates data extraction, reduces errors, and speeds up accounts payable with LlamaParse.
With our UK team expanding by more than 200% in six months, we’ve opened a new office at The Gilbert in Finsbury Square and brought in senior leaders to…
Discover how agentic OCR transforms document processing with multimodal reasoning, self-correction loops, and template-free automation.
What forward deployed engineering means at OpenAI
Learn how OCR for accounts payable automates invoice processing, improves accuracy, reduces costs, and integrates structured data into ERP systems.
Fifteen years ago, TLS was something you'd use to protect the most sensitive areas of your app. Then Google, Let's Encrypt, the EFF, and others led the push for HTTPS everywhere, making encryption the default for every request. Without…
Connect Cursor to Langfuse to trace every agent turn, pinpoint expensive model and tool calls, and use evaluations to confirm a fix actually helps, with ClickHouse powering the analytics underneath.
I have unkind things to say about people who make their own pasta. Here, I’ll only use them as an example of how not to think when you’re building infrastructure. It’s unfairly easy to believe that the harder you work on something, the…
Learn how to build an AI agent once and reuse it across IT, HR, and ops teams, with shared governance, identity, and observability.
Uber is rolling out ChatGPT Work to more than 30,000 employees, helping teams research, analyze data, create presentations, and complete complex workflows.
You can now deploy apps to Vercel without an account using Vercel CLI . Share prototypes faster with a live URL before signing in to claim your project. From a project directory that isn't linked to Vercel, run: The CLI builds your…
Learn how to connect on-premises data sources to Amazon Quick privately with AWS PrivateLink, keeping analytics traffic off the public internet. This walkthrough covers VPC endpoints for user and API access, VPC connections for data…
Many teams build a multi-agent AI prototype in days but hit a wall in production. This post defines five production challenges (orchestration, per-agent model selection, reliable operation, secure code execution, and monitoring) and…
The Adaptive Agentic Intelligence Framework (AAIF), built with Strands Agents, handles routine tasks locally with small language models and routes complex requests to cloud models through Amazon Bedrock. This post walks through building…
How OpenAI works with distribution partners to uphold consistent safeguards, support responsible access, and clarify expectations in our Usage Policies.
Train Qwen3 on AWS Trainium with torchtitan three ways on a single trn2.48xlarge: the Hugging Face Transformers v5 modeling backend, torchtitan's native Qwen3, and NKI mixture-of-experts kernels. The same torchrun launcher and config…
U.S. Hunger, a nonprofit addressing food insecurity, uses Amazon Quick to connect organizational data with leadership decisions. Leaders explore applicant data directly, share KPI definitions and financial context in Quick Spaces, and…
Today's PHP Security Release, and How We're Strengthening PHP Security If you're running anything on PHP it's vital you upgrade - today's updates (across the entire PHP 8.x line) address two remote code execution holes, and there have…
OpenReplay replaced Postgres with ClickHouse to ingest billions of events and up to 6 TB a day on a single, vertically scaled instance, with sub-second queries and retention handled by a single TTL setting.
The Agents CLI Extension System enables developers to use frameworks like LangChain, LangGraph, or custom enterprise stacks while retaining a unified deployment and evaluation workflow. It details a first-party LangChain template that…
You can now write routing middleware in Python for FastAPI, Flask, Django, and other Python frameworks. It runs before requests reach your app, and can rewrite, redirect, or answer them. To set it up, point the proxy entrypoint in…
AT&T is working with OpenAI to make legal knowledge easier to access. Learn how its LegalEdge team is rethinking intake and scaling legal judgment.
mammoth-1 and elephant-1 are decision models trained by our internal research team, Base Labs. They are open-source and free to try for a week.
Panache — RTS Belle Époque pour mobile Well this is delightful: I released Battle Drones yesterday under an Apache 2.0 License and there's already a French language fork which reskins the game to take place in steampunk 1900s France and…
We moved a copy of 99.99999% of GitHub's repositories to Azure. Here's how per-reference quorums will bring over the rest, and how object storage lets reads scale independently. The post The bridge to agent-scale Git infrastructure…
Cloudflare's web application firewall can now correlate an HTTP request with its response. Full transaction detection helps every customer find successful exploits, data exposure, and risky misconfigurations that request-only inspection…
Cloudflare has deployed default web application firewall (WAF) protections for newly disclosed PHP vulnerabilities before public disclosure. The rules use request and response signals to stop exploit attempts and prevent exposed data…
NVIDIA Vera tops the ClickBench hardware leaderboard, and three micro-benchmarks for filters, aggregation and joins show ClickHouse scaling almost linearly across its 176 cores.
Claude Mods is an early form of the Attention Interface. It flips the agent harness paradigm from the harness facing the machine, to the harness facing the human.
Claude Mods is an early form of the Attention Interface. It flips the agent harness paradigm from the harness facing the machine, to the harness facing the human.
Almost 100,000 Commerce.com domains now give AI agents real shopping tools through WebMCP on Cloudflare. Here's how we turned it on across every store at once, while keeping each merchant in control.
From local businesses to global brands, more advertisers can now reach people as they explore, compare and decide what to buy in ChatGPT.
Protein design is a search problem. Choose amino acids for an eight-residue protein sequence and more than 25 billion combinations are possible. Choose amino...
OpenAI and Google Cloud are bringing frontier models and Codex workflows to Gemini’s Model Garden, available through Google Cloud Marketplace.
Google Cloud has always been defined by choice, enterprise-grade governance, and market-leading price-performance. Key to the foundation of openness and choice is giving customers flexibility to run agentic workloads across their choice…
Reinforcement learning for LLMs is hard, so we built Modal Dojo — a library that gives you the building blocks, tuned defaults, and observability to get started with production-grade LLM training.
An AI agent can complete its assignment and still cross a security boundary. Recent incidents show that agents can circumvent security controls while using...
Autonomous agents can work for hours on complex tasks, accessing files, tools, the internet, and spawning other agents to collaborate with. Executing complex...
Building an AI factory digital twin requires developers to bring site plans, equipment assets, and engineering requirements into a consistent model that teams...
AI factories turn energy into intelligence. But before an AI workload can produce its first token, operators must bring an interconnected system of compute,...
Learn how OpenAI built ChatGPT’s LLM-powered ad stack to understand conversational intent, keep ads separate from answers, and deliver relevant ads at scale.
Patch the Planet connects security researchers, maintainers, and incident responders to help address critical vulnerabilities across open source.
An open source “bomb” drops from China
An open source “bomb” drops from China
Listen now (44 mins) | Paying for outcomes, building with agents, and the last mile of enterprise AI adoption
Paying for outcomes, building with agents, and the last mile of enterprise AI adoption
See how OpenAI, PerturbAI, and Scripps Research used distributed AI agents to uncover promising research leads in data from millions of brain cells.
1350 older posts here are not shown; this tab lists the 150 most recent, and every post is still under its own topic. Narrow the window to bring them within the cap.
How an agent loop is built, driven, and kept on task.
An agent working on production systems needs permission to act. A request like “fix this deployment” doesn’t tell you upfront which actions it will take or what access it will need. Grant agents broad access and a mistake can affect…
AI agents now write a growing share of production data pipelines (over 70% of SDP...
Bayer Crop Science built the Agentic Decision Engine (ADE), a multi-agent platform on Amazon Bedrock AgentCore that lets breeders, analysts, and researchers ask natural language questions about their analytics platform. This post covers…
Teaching a robot to navigate an obstacle course requires more than a walking policy. It must perceive its surroundings, maintain balance, and adjust its actions...
Kimi K3.1 Pro from Moonshot AI is now available on AI Gateway , along with Kimi K3.1 Pro Highspeed , a faster version for interactive use. K3.1 Pro is the first model in Moonshot's K3.1 series. It builds on Kimi K3 with improvements in…
Voice agents often wait for an agent's full response before starting speech synthesis, adding latency users feel directly. This post shows how to stream each sentence to a Kokoro-82M text-to-speech model on Amazon SageMaker AI as the…
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…
Rescale used the AI-Driven Development Lifecycle (AI-DLC) framework on AWS, with Strands Agents and Amazon Bedrock, to build an intelligent test-triage agent in weeks. Learn how AI-DLC structures AI-assisted development and how Rescale…
LangSmith LLM Gateway is in public beta: spend caps, rate limits, model fallbacks and PII redaction for production agents, without provider lock-in.
LlamaParse is the most accurate agentic OCR Platform. Parse and Extract your complex documents at the Pareto frontier of cost and accuracy.
Learn why standard OCR falls short for KYC compliance and how agentic document extraction delivers the field-level accuracy AML regulations require.
Learn why single-pass extraction fails and how deep extraction uses agentic verification to deliver production-grade accuracy on complex documents.
Agentic document extraction uses AI reasoning and visual grounding to accurately process complex documents without templates. Learn how it works.
Agentic document processing uses AI agents to autonomously handle document workflows end to end. Learn how it works and where to start.
OCR for receipts breaks when layouts vary and rules pile up. Discover how agentic OCR reconstructs line items, totals, and structured data for automation.
Discover how agentic OCR transforms document processing with multimodal reasoning, self-correction loops, and template-free automation.
Many teams build a multi-agent AI prototype in days but hit a wall in production. This post defines five production challenges (orchestration, per-agent model selection, reliable operation, secure code execution, and monitoring) and…
The Adaptive Agentic Intelligence Framework (AAIF), built with Strands Agents, handles routine tasks locally with small language models and routes complex requests to cloud models through Amazon Bedrock. This post walks through building…
The Agents CLI Extension System enables developers to use frameworks like LangChain, LangGraph, or custom enterprise stacks while retaining a unified deployment and evaluation workflow. It details a first-party LangChain template that…
Claude Mods is an early form of the Attention Interface. It flips the agent harness paradigm from the harness facing the machine, to the harness facing the human.
Claude Mods is an early form of the Attention Interface. It flips the agent harness paradigm from the harness facing the machine, to the harness facing the human.
Almost 100,000 Commerce.com domains now give AI agents real shopping tools through WebMCP on Cloudflare. Here's how we turned it on across every store at once, while keeping each merchant in control.
Google Cloud has always been defined by choice, enterprise-grade governance, and market-leading price-performance. Key to the foundation of openness and choice is giving customers flexibility to run agentic workloads across their choice…
An AI agent can complete its assignment and still cross a security boundary. Recent incidents show that agents can circumvent security controls while using...
Autonomous agents can work for hours on complex tasks, accessing files, tools, the internet, and spawning other agents to collaborate with. Executing complex...
Building an AI factory digital twin requires developers to bring site plans, equipment assets, and engineering requirements into a consistent model that teams...
See how OpenAI, PerturbAI, and Scripps Research used distributed AI agents to uncover promising research leads in data from millions of brain cells.
Learn how to secure multi-agent AI systems with cryptographic identity using Keyfactor EJBCA and AWS services. This reference architecture combines signed request envelopes, user-signed delegation assertions, RFC 8693 audience-scoped…
Build a governed semantic layer for agentic AI on AWS with the Denodo Platform and Amazon Bedrock AgentCore. It federates Amazon Aurora, Amazon Redshift, and an on-premises PostgreSQL database into a single customer 360 view an agent…
The third post in our semantic-layer series builds a customer 360 across Amazon Redshift and Amazon Aurora with Strategy Mosaic, exposes it to a Strands agent on Amazon Bedrock AgentCore through the Mosaic MCP server, and enforces…
The third post in our semantic-layer series builds a governed customer 360 model on Amazon Redshift with AtScale, then exposes it to a Strands agent on Amazon Bedrock AgentCore through the AtScale MCP server. Identity propagation, RBAC,…
Learn about the top reasons agentic AI projects stall before reaching production, and how to get your data and your workforce AI-ready.
Your AI agent is only as good as its harness. Learn why better models alone aren’t enough and what it takes to build more effective AI agents.
The open-source Agent Development Kit (ADK) and gemini-3.8-live enable developers to build responsive voice agents by implementing non-blocking tools in Python. Developers can use streaming tools to yield progress updates or background…
We benchmarked NVIDIA's new Arm-based Vera CPU on agentic Sandbox workloads. Under heavy load, it completes far more tasks than the x86 and Graviton alternatives, without sacrificing latency.
Developers are familiar with running coding agents locally through harnesses like Pi, Claude Code, and others. They are reliable daily drivers for generating code, but keeping agents running around the clock requires your computer to be…
As organizations work to leverage AI to transform their businesses, it is clear that the raw model intelligence is not the key challenge. Consistently, we hear that the challenges lie in how the agents can be deployed across the company…
DeepSeek V4.1 Flash 1017 is now available on AI Gateway . DeepSeek V4.1 Flash 1017 is a minor update to DeepSeek V4.1 Flash that improves reasoning, coding, agentic tasks, and image understanding. The coding gains show up in terminal…
In production systems today, most AI agents operate as lightweight dispatchers. They parse a prompt, make an API call, retrieve a record, and hand back control. Their execution is fast, stateless and cheap. Biotechnology does not work…
Reflection AI's Beam is an open-weight model for agentic coding and long-horizon tasks. This post shows how to deploy Beam on AWS: start with an Amazon SageMaker JumpStart real-time endpoint, or use Amazon SageMaker HyperPod Inference…
Beam is now available on Baseten. Reflection’s 501B open-weight MoE model is built for efficient reasoning in coding and agentic workflows, with 1M context.
Agents already shop on the web, but they often use pages and tools originally built for people and search engines. Building a better shopping experience for agents means treating them as first-class shoppers, with product information…
Nick Cammarata out of nowhere with some major claims!
Nick Cammarata out of nowhere with some major claims!
Two engineers shipped Stack Overflow for Agents in a few weeks with Codex, and its ChatGPT plugin helped drive up to 10× growth in agent creation.
Google OAuth Platform now supports DPoP (Demonstrating Proof-of-Possession) to sender-constrain refresh tokens, neutralizing Infostealer malware and adversary-in-the-middle attacks. By cryptographically binding long-lived refresh tokens…
Meta’s newest app, Forum, is for anyone who wants to go deeper into the conversations in their Facebook groups. To build Forum, Meta engineers adopted a whole new way of working by bringing an AI agent into the development cycle. We’re…
Meta launches Muse Spark Open, a coding-optimized 816B mixture-of-experts (MoE) model with 42B active parameters and a 1M token context window. Trained on...
Muse Spark Open is now available on Baseten. This 816B open-weight omnimodal model is optimized for coding agent workflows, with a 1M token context.
Small teams, short runway, and constant pressure to ship faster — it’s the reality we see every day among the founders we work with at Google for Startups. When you are building a company from the ground up, every hour spent on…
For a decade, Vercel has shaped how the web is built. Now, we’re doing the same for agents. The companies that win the next decade will build on infrastructure designed for agents from the start, and thousands of people gathered across…
Enterprise leaders want AI agents to take on more work but don't fully trust them to act correctly every time. Trust in Depth is AWS's layered approach-scope, know, bound, observe, and prove-that lets agents act on their own when their…
Always-on agents are taking on longer, more complex enterprise workflows. A single request can involve reading a document, checking information across sources,...
Agents often handle complex work that continues across runs. As that work evolves, they may need a new environment with different tools, runtimes, or compute. They should be able to continue working with the same files when they change…
Cat Jam is now available on AI Gateway as a stealth model, free to use during the stealth period. Cat Jam is a reasoning model for coding and agentic tasks, with low-latency inference. It supports function calling , so agents can…
Muse Spark Open from Meta is now available on AI Gateway. The open-weight reasoning model is designed for coding and agentic workflows, including multi-file refactoring, debugging, and understanding large codebases. Muse Spark Open…
At AI Engineer New York, frontier labs and financial firms explored how to give agents useful autonomy while keeping consequential decisions under control.
At AI Engineer New York, frontier labs and financial firms explored how to give agents useful autonomy while keeping consequential decisions under control.
Learn how to build an agentic intelligent document processing (IDP) system using Amazon Bedrock AgentCore that classifies, extracts, validates, and learns from business documents with minimal human intervention, automatically handling…
See how Amp pushes the frontier of what agents can do with GPT-6 Astra, Ultrafast, GPT-Live-1, and Sign in with ChatGPT.
New analysis shows how agentic AI adoption is growing across EU enterprises and small businesses—and where deeper use could unlock more value.
Our first robotics pod with Dyna’s Chief Scientist on their big bet on World Action Models & the Next Frontier of Physical AI
Our first robotics pod with Dyna’s Chief Scientist on their big bet on World Action Models & the Next Frontier of Physical AI
Enterprise applications have become increasingly demanding. Agentic AI has compounded the need for vast amounts of enterprise data with frequent, complex...
Learn how the AWS Generative AI Innovation Center Partner Agent Factory and Anchor Browser built AI agents that autonomously pay for x402-protected web resources and APIs. Amazon Bedrock AgentCore payments checks budgets, signs…
Built on Deep Agents and LangGraph, Open SWE provides the core architectural components for internal coding agents.
Today we’re launching Deep Agents deploy in beta. Deep Agents deploy is the fastest way to deploy a model agnostic, open source agent harness in a production ready way.
Harness engineering improved LangChain's coding agent from Top 30 to Top 5 on Terminal Bench using self-verification, tracing, and context optimization.
Grok 4.7 is now available on Amazon Bedrock. In this post, we build a tool-calling agent on Grok 4.7 and evaluate it on three reasoning tasks at four reasoning effort levels, measuring correctness and token cost per completed task. We…
If the agent isn't using my diagnostics library, I could introspect and update AGENTS.md, but there's still a delta, and it'll drift. Instead, I write a pre-commit hook that checks the git diff for direct calls to observability modules…
AI Search in LangSmith is a new way to query traces using natural language. Describe the behavior you're looking for, and LangSmith returns the traces where it happened.
Use ScaleDown's constrained decoding models with LangChain and Deep Agents to summarize history, compress tool results, and extract data faster and cheaper.
At AI Engineer New York, Anthropic drew on lessons from Unix and Linux to explain how agents can tackle bigger projects without overwhelming developers.
At AI Engineer New York, Anthropic drew on lessons from Unix and Linux to explain how agents can tackle bigger projects without overwhelming developers.
Your agent can now use the Vercel CLI to buy a domain. The agent discovers candidates with vercel domains search , checks availability and registrar pricing with vercel domains check and vercel domains price , and initiates the purchase…
Managed Deep Agents includes a new API for managing reactions for your distributed agents, and a system to dynamically assign emoji responses with your instrument of choice. Learn more.
Building an AI agent that works in a demo is a different problem from running one for 40 million developers. Postman and AWS share the architectural patterns behind Agent Mode: controlling tool sprawl, exposing schema-based reads, and…
Rillet on Vercel Tripled shipping rate in three months with eve agents deployed on Vercel Two-person team closed 800+ Linear tickets in its first 4 months Builder access via Enterprise Managed Users, SSO, and Directory Sync Ships…
Discover how Snyk transformed an internal support agent into Snyk Assist, a customer-facing AI feature powered by LangChain, LangGraph, and LangSmith.
Amazon Bedrock AgentCore payments gives AI agents a managed way to pay for services on demand, with spending limits enforced by the infrastructure. See how Incarna's agents pay BlockRun for model inference one request at a time over…
The NVIDIA KGMON team placed second in the KDD Cup 2026 Data Agents competition with a system built around a simple idea of making an agent's harness smaller,...
Build an agent that pays for real purchases. Restock runs in Slack on Managed Deep Agents and pays with Stripe's Link over the Machine Payments Protocol.
In recent decades, Ireland has grown into a vibrant hub for global technology. As modernization accelerates, Irish organizations — spanning government agencies, established enterprise brands, and growing startups in AI and other areas —…
LangChain just released new capabilities in Managed Deep Agents. Agents can now schedule follow-ups, reconfigure themselves on every run, and react to Slack messages.
Cloudflare Managed Defense uses a team of specialized AI agents built on Workers and global network telemetry to analyze security alerts. By separating deterministic evidence collection from model inference, the system delivers grounded…
Discover how to automate loan processing workflows using AI document extraction. Learn how to extract key data from W-2s, bank statements, and tax returns to speed up approvals.
AI factories are some of the most complex operations in the world, combining GPUs, CPUs, switches, DPUs, and SuperNICs alongside schedulers, orchestration...
The RPA-era ROI model misses most of the value agentic automation creates. This post gives AI center of excellence leaders a framework to size the full value of agents across time savings, exception handling, decision quality, and…
Qlik built Qlik Answers on Amazon Bedrock to give its 40,000+ customers grounded, sourced answers across structured and unstructured enterprise data. Learn how a layered, multi-agent architecture with cross-Region inference and Amazon…
AWS DevOps Agent can diagnose production incidents but is kept in observe-and-report mode so it does not change resources directly. This post shows how to use AWS Lambda Durable Functions, Amazon EventBridge, and Amazon Bedrock to turn…
Cornerstone OnDemand built Orion AI, a multi-agent system on Amazon Bedrock and Strands Agents, to turn database operations from reactive firefighting into proactive automation. A three-person team cut database diagnosis from 45 minutes…
LlamaParse is the most accurate agentic OCR Platform. Parse and Extract your complex documents at the Pareto frontier of cost and accuracy.
Kubernetes co-creators Craig McLuckie and Joe Beda aim to bring agent harnesses fully into the cloud.
OpenAI “rogue” agent activities found on Wikimedia projects Given how tempting a target wikis are for rogue agent swarms, it's not a huge surprise that Wikipedia found evidence of that activity once they went looking: The Wikimedia…
We're rebuilding GitHub's Git infrastructure while GitHub keeps running, creating a foundation for agent-scale software development. The post Building Git infrastructure for agent-scale development appeared first on The GitHub Blog .
See how Redpanda powers real-time data streaming and agentic AI governance for global manufacturers.
GLM 5.3 from Z.ai is now available on Amazon Bedrock: a 753B-parameter mixture-of-experts model built for coding and long-horizon agentic tasks. Learn how to invoke it with the OpenAI-compatible APIs, cut cost and latency with prompt…
Anthropic Claude Opus 5.5 and Claude Sonnet 5.5 are available on Amazon Bedrock in the AWS GovCloud (US) Regions. Learn how to use them with Claude Code, Anthropic's agentic coding tool, for compliance-aligned, AI-assisted development…
126 older posts in this topic are not shown; the page lists the 100 most recent of each topic. Narrow the window to bring them within the cap.
Agents that read, write, and review code, and the benchmarks that grade them.
Connect Cursor to Langfuse to trace every agent turn, pinpoint expensive model and tool calls, and use evaluations to confirm a fix actually helps, with ClickHouse powering the analytics underneath.
OpenAI and Google Cloud are bringing frontier models and Codex workflows to Gemini’s Model Garden, available through Google Cloud Marketplace.
OpenHands adds day-zero support for Recursive Flint, an efficient open-weight coding model built for agentic workflows and fast local inference.
Cloudflare uses OpenAI Daybreak to investigate vulnerabilities and propose fixes for review. Its developers use Codex to build faster and improve code quality.
Three Java Champions share how they use GitHub Copilot to solve problems, accelerate development, and keep humans responsible for the results. The post How 3 Java Champions are using GitHub Copilot for increased productivity appeared…
Token Pass now includes DeepSeek V4.1 Flash, alongside Kimi K3 and GLM 5.3, each accessed through its preferred provider, with more models to come. The DeepSeek model uses the latest weights (1017). Token Pass is a $30/month…
My obituary for programming Jeremy Scheff poignantly captures the key loss caused by coding agents for career programmers who enjoyed the puzzle of developing code: Programming turned into one of the greatest jobs you could imagine. You…
Learn how to move an agent session from the GitHub Copilot app into VS Code without losing the conversation, code changes, or working context. The post How to continue a GitHub Copilot app session in VS Code appeared first on The GitHub…
Learn how OpenAI is improving ChatGPT across low-resource programming languages, with new evaluations for software engineering, SQLite, and COBOL migration.
Anant Kumar builds tools to compare loan valuations and analyze credit agreements, using Codex for code review and faster follow-ups.
AI adoption in the enterprise is growing rapidly, and expanding beyond engineering users. In the last three months, token usage by enterprise companies on AI Gateway has quadrupled, and OpenAI recently reported that from August 2025 to…
Run AI coding agents on DigitalOcean with OpenHands. Access managed cloud infrastructure, configure models, and manage agent sessions from Agent Canvas.
Learn how to add custom instructions and skills to your repository, use custom agents, and connect MCP servers like Playwright. The post GitHub Copilot app for Beginners: How to teach Copilot about your codebase appeared first on The…
Research: Profiling web-edit with Python 3.15: flame graphs and source heatmaps I put the new Python 3.15 profiler through its paces by having GPT-6 Astra Extra High in Codex run it against my new web-edit app : I need a Python 3.15…
Release: web-edit 0.1a0 I was working on a project from my phone using Codex Remote when I found myself wanting to edit a credentials file without round-tripping those credentials through the LLM itself. So... I had Codex and GPT-6…
I shipped a new feature for my blog today: the Newsletters page, which offers an index of all of the newsletters I've sent out, both my free weekly Substack and my monthly sponsors-only updates. I built the feature almost entirely using…
Using GPT-6 Astra in Codex, Asana made its browser agent 76x cheaper and 5x faster in tests to offer customers more capable models.
Across recruiting, engineering, and operations, Oracle turns specialist knowledge into fast, repeatable workflows with ChatGPT Work and Codex.
LegalOn cut estimated daily Codex costs by 65% while maintaining development speed. It matched Astra, Sol, and Luna to tasks and managed budgets strategically.
Step 5 Preview , StepFun's flagship model, is now available on AI Gateway . It's built for agentic coding, professional knowledge work, and financial analysis, from building apps and debugging code to research and analytical reports.…
OpenHands supports NVIDIA Nemotron 3.5 Super VL, bringing open multimodal AI models to always-on agent workflows for software engineering teams.
OpenHands Agent Canvas added scoped Agent Profiles, persistent automations, isolated Docker runtimes, and easier Apps in September 2026.
No partnership, no custom API. Just an auth key and a tailnet.
Mistral Large 4 is now available on AI Gateway. The open-weight, natively multimodal model combines reasoning with text and image understanding. It is useful for cyber defense, manufacturing, and finance workflows. You can access the…
Amazon SageMaker optimized generative AI inference introduces the aws-ai-ml skill through the Agent Toolkit for AWS, giving coding agents like Kiro, Claude Code, and Codex deep expertise in inference optimization and benchmarking.…
Baseten is partnering with OpenAI to serve open models natively via Codex and the Responses API.
Explore more than 20 announcements from OpenAI DevDay 2026, including GPT-6 Astra, ChatGPT, Codex, APIs, security, and new tools for builders.
With Codex, GPT-Live-1, and GPT-6 Astra, Proaction builds, operates, and sells modern fleet management faster.
Describe the interface you need in plain English, then let the agent build a live surface you can both use and update—so you spend less time adapting to tools and more time getting work done. The post GitHub Copilot app for Beginners:…
The latest OpenHands Enterprise release is designed to help organizations scale their agentic SDLC with better usability, governance, and visibility.
How we rebuilt the diff surface in the GitHub Copilot app to open a million-line pull request with hundreds of inline review comments. The post Rendering huge pull requests in the GitHub Copilot app appeared first on The GitHub Blog .
Pair OpenCode, an open-source terminal-native AI coding agent, with open weight models on Amazon Bedrock to get a secure, flexible, pay-per-use coding assistant. Learn how to configure multi-model workflows, match the right model to…
Inside a global bank's shift to self-serve dedicated inference: how Together's DMI gave engineering teams direct control over scaling, models, and testing.
You and your agents can now deploy static artifacts to Vercel in under one second through Vercel CLI. Run vercel deploy to share a prototype, publish an HTML report, or preview a page created by your coding agent. Vercel automatically…
A rewrite this size wasn't affordable before agents. Here's what porting the Copilot agent runtime to 800,000 lines of production Rust actually took. The post Migrating the GitHub Copilot runtime to Rust, using Copilot appeared first on…
Learn how ChatGPT Work and Codex analytics help teams understand AI usage and spend, identify training needs, and connect adoption to business outcomes.
My comment on Feeling sad about AI — Hacker News. I'm not sure how useful it is to say this, but I think a lot of people (myself included, a few years ago now) have been through this moment of existential crisis and come out the other…
Fusion is the most efficient frontier harness for Fable and Astra, up to 39% more efficient compared to other model harnesses across major coding…
Fusion is the most efficient frontier harness for Fable and Astra, up to 39% more efficient compared to other model harnesses across major coding…
GPT‑6 Astra improves Devin’s ability to test software and show that it works, with the goal of helping engineers review less code and ship more.
Checking agent-generated code usually means hopping between tabs. Learn how to view diffs, run terminal commands, and preview web apps side by side in the GitHub Copilot app. The post GitHub Copilot app for Beginners: Using the diff,…
The AI SDK harness layer now supports GitHub Copilot through the official @ai-sdk/harness-github-copilot adapter. The harness layer lets your application run different coding agents through the same HarnessAgent interface, so you can…
César de la Fuente’s lab uses Codex and ChatGPT to search living and extinct genomes for antimicrobial candidates to fight drug-resistant infections.
See how Credit Genie uses OpenWiki to automate repo documentation, reduce tribal knowledge, and give engineers and coding agents searchable codebase context.
Tool: .blend URL Viewer I'm continuing to have a lot of fun with GPT-6 Astra and Blender (see my TIL ). As a big fan of the Imperial Fabergé Easter eggs , I've always thought it would be fun to make some new ones that celebrate popular…
How Devin and a Cognition researcher built the world’s highest-performance GPU lattice siever, to make factoring numbers 10x cheaper than the previous…
Cognition has raised over $2B at a $48B valuation, led by Andreessen Horowitz and Accel, to build the future of software engineering.
See how an MIT researcher uses GPT-5.6 Sol with Codex to autonomously run quantum computing experiments, analyze results, and calibrate qubits.
GPT 6 Astra from OpenAI is now available on AI Gateway. GPT 6 Astra is designed for long-running agentic work. It can navigate software, analyze data, run simulations, and build and test websites while adapting to new requirements,…
Deploy a customer-operated LiteLLM gateway on Amazon ECS with AWS Fargate, connect it to an OpenAI model on Amazon Bedrock, and configure Codex to route requests through the gateway's Responses API with scoped identities, budgets, rate…
Learn how to run parallel agents in the GitHub Copilot app, and experience the moment it stops feeling scary and starts feeling powerful. The post GitHub Copilot app for Beginners: Run several agents at once appeared first on The GitHub…
Tool interfaces, protocols, memory, and what the model is given to work with.
Vercel MCP now includes submit_feedback , a tool for sending bug reports, friction reports, and feature requests directly to the Vercel team from your agent or harness. Ask your agent to give feedback in plain language: Send Vercel…
Agents waste time and context on clumsy searches, oversized results and stale data. Learn the retrieval techniques LangChain's Deep Agents harness uses to get the right context in.
Vercel Connect can now create a separate restricted Stripe API key for each agent or task. When you create a Stripe connector, you can now choose Managed API Key as the connection method. Vercel creates the key in your Stripe account,…
In Databricks’ State of AI Agents 2026 report, there’s a chart from Neon’s telemetry that’s somehow both mind-blowing and unsurprising. In October 2023, AI agents created 0.1% of the databases on Neon. By October 2025 they created 80%…
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)…
Learn how IskayPet, a leading pet care company in Iberia, built a multilingual corporate knowledge assistant on Amazon Bedrock Managed Knowledge Bases. The assistant answers employee questions across retail, veterinary, onboarding, and…
As Cybersecurity Awareness Month continues, the GitHub Bug Bounty team spotlights @r31n, diving into their research approach, favorite bug classes, and advice for aspiring hackers. The post How bug bounty sharpens one researcher’s…
Learn how Cerebras redesigned engineering interviews for the AI era, evaluating AI collaboration, engineering judgment, verification, and real-world skills.
Mammoth-1 and Elephant-1 from Baseten are now available for free on AI Gateway for one week. Free access is available through October 27. These decision models classify text, route requests, and score content against criteria you…
The Smartsheet Model Context Protocol (MCP) Server lets AI agents read, update, and act on live data. This post shows two ways to connect it on AWS: Amazon Quick for business teams and Kiro for developers, so your agents can work with…
The Smartsheet Model Context Protocol (MCP) Server lets AI agents read, update, and act on live data. This post shows two ways to connect it on AWS: Amazon Quick for business teams and Kiro for developers, so your agents can work with…
Hardcoded tool wiring breaks down as an agent's catalog grows from a handful of tools to hundreds. This post walks through a runtime Discover, Instantiate, Execute loop built on AWS Agent Registry, where an orchestrator discovers MCP…
Learn how Eon replaced per-tool credentials and local MCP processes across AWS, GCP, and Azure with a single authenticated Amazon Bedrock AgentCore gateway, giving more than 100 developers and autonomous agents unified access to…
Earlier this month at Gemini at Work, we introduced the Gemini agent: a single, universal agent for work. It answers your questions, handles your knowledge work, creates your images and media, and writes and runs code. You give it…
MiniMax M3.1 Flash from MiniMax is now available on AI Gateway for free until Oct 22. MiniMax M3.1 Flash is an open-weight multimodal model for coding and agentic work, with a 1M-token context window for long documents and multi-step…
Release: llm-tools-datasette 0.1.1 When I switched LLM over to httpx2 in LLM 0.33 I forget that this would break any plugins which used httpx without directly depending on it. llm-tools-datasette was one of those plugins, now fixed.…
Beam from Reflection AI , previously available in stealth as stealth/cat-jam , is now available on AI Gateway . Beam is free to use for a limited time during the launch period when serviced by Reflection AI. Beam is an low latency US…
The google/adk-recipes repository unifies Agent Development Kit (ADK) resources into three structured layers—core/, plugins/, and contrib/—serving as a single source of truth for building AI agents. Developers can leverage reference…
Vercel Connect now supports MCP Events in public beta, allowing your apps and agents to subscribe to webhook events from compatible MCP servers. Build agents for tasks like summarizing new messages, reviewing newly created issues, or…
Meta Muse Spark Open, Meta's open-weight model for agentic coding, is live on Together AI with a 1M-token context window on serverless and dedicated infrastructure.
Apigee hybrid 1.17.0 introduces native Model Context Protocol (MCP) support, enabling the MCP data plane to run directly within your own Kubernetes cluster. This architecture ensures that AI agent request traffic and tool execution…
Long-running agents degrade as context piles up. Learn how offloading, summarization, skills, subagents, and memory keep your agent's context window focused.
This post walks you through building an AI agent system with Amazon Bedrock AgentCore that automates PowerPoint presentation creation from enterprise documents. A supervisor agent, Retrieval Augmented Generation, and Amazon Bedrock…
Open models like GLM-5 and MiniMax M2.7 now match closed frontier models on core agent tasks — file operations, tool use, and instruction following — at a fraction of the cost and latency. Here's what our evals show and how to start…
How our engineering team ran Gemini 4 Argon unattended overnight with .agents/skills and /goal mode to cut in-memory session prep overhead by 40% in long sessions and speed up cold imports by 21% in Google's open-source Python Agent…
Google is launching the Developer Knowledge API and MCP Server in public preview. This new toolset provides a canonical, machine-readable way for AI assistants and agentic platforms to search and retrieve up-to-date documentation across…
Deep Agents now lets you bind tools to skills, pin skills at runtime, and reload skills mid-thread, so agents with expansive skill repositories stay context-efficient and effective.
Enterprise RAG unlocks insights from knowledge sources like SharePoint, Google Drive, and Confluence, but those sources carry complex permissions. Learn how Amazon Quick and Amazon Bedrock Knowledge Bases enforce document-level access…
Early access to Glyph Cluster is now available on AI Gateway for teams on a Pro or Enterprise plan with purchased AI Gateway credits. Usage is free during the stealth period. Glyph Cluster is a reasoning model for coding and…
Off-the-shelf AI assistants forget you between conversations. This post shows how to build a personal assistant that accumulates context using OpenClaw on Amazon Bedrock AgentCore runtime, with AgentCore memory turning disposable chats…
EmbeddingGemma 2 is a compact, open-source multimodal embedding model that maps text, code, images, video, and audio into a unified 768-dimensional space. Developers can use the sentence-transformers library to selectively load modular…
Add a voice travel concierge to an airline app with Amazon Bedrock AgentCore, Amazon Nova Sonic for real-time speech, and Amazon Bedrock Knowledge Bases for policy answers. Travelers change seats, check delays, and ask policy questions…
Learn how OpenAI and Ironclad are training and evaluating AI agents on complex contracting workflows to advance computer use for professional work.
Learn how startups can choose GPT-6 models, tune reasoning effort, improve prompts and skills, coordinate tools, and prepare workflows for production.
The Adjudicated Query pattern pairs the Amazon Quick chat agent with a bounded MCP server over a deterministic rules engine to deliver provably complete, defensible compliance answers. This post walks through the reference architecture…
Learn to direct AI agents, critically review their output, and keep technical judgment at the center of your workflow. The post AI is changing developer work. Here are three skills to strengthen. appeared first on The GitHub Blog .
AI agents are becoming a standard part of development workflows, but general-purpose agents weren't built with specialized infrastructure software such as...
Learn how to use Amazon S3 Vectors as the persistent memory layer within the NVIDIA NeMo Agent Toolkit (NAT), deployed on Amazon Elastic Kubernetes Service (Amazon EKS). This post shows how NAT's memory subsystem works and how to…
Our DevDay coverage - the first pod on the DevDay lineup - dives in with the leaders of OpenAI’s CUA team and API platform.
Browserbase Search and Fetch are now available through AI Gateway . Give your models access to current information, find relevant pages, and retrieve their contents, alongside Browserbase's existing AI SDK and Vercel Marketplace…
Today, we’re expanding our ecosystem of managed remote MCP servers by introducing the Google Cloud CLI remote MCP server in preview. Powered by the popular gcloud and bq (BigQuery) command-line tools, this new server gives AI agents…
Cloudflare’s AI Gateway, Ceramic.ai, Stocktwits, and more are using the Cloudflare Monetization Gateway today to charge agents for access to tokens, APIs, and MCP tools. U.S.-based sellers can now apply for access to the closed beta.
Ling 3.1 Flash from InclusionAI is now available on AI Gateway. The model is free to use through October 13, 2026. Ling 3.1 Flash is a hybrid reasoning language model with 560B total parameters and 25B active per token. It has a…
GPT-6.1 Sol is now generally available on Amazon Bedrock, bringing stronger reasoning to coding, computer use, and professional workloads that run frequently.
Prompt engineering in Amazon Quick shapes how accurately its AI-powered features respond to your requests. Part 1 of a two-part series covers the foundational principles and reusable frameworks (specificity, context-setting, few-shot…
Part 2 of our Amazon Quick prompt engineering series goes component by component. Learn the prompt patterns that get the best results from Amazon Quick Research, Quick Flows, Quick Sight, chat agents, and action integrations, plus the…
Manually extracting data from hundreds of vendor contracts doesn't scale, and RAG chat tools fall short on portfolio-wide questions. This post shares a contract intelligence platform on AWS that uses AI agents to extract and verify…
We are expanding access to Cloudforce One's Threat Events Platform to every Cloudflare account and introducing Threat Signals. Threat Signals automatically parses open-source threat reporting, extracts structured indicators, and…
Meet GPT-6.1 Sol: near-Astra intelligence for coding, computer use, and professional work at one-fifth of Astra’s standard API input and output token prices.
GPT-6.1 Sol from OpenAI is now available on AI Gateway. It improves on GPT-6 Sol for coding, computer use, and professional work, including reading complex documents and carrying out multi-step workflows. The model is suited to agents…
xAI's Grok 4.7 is now available on Amazon Bedrock: a frontier model for coding, long-running agents, and knowledge work. It offers a 500K token context window and four configurable reasoning effort levels, reachable through the…
Traditional OLTP systems weren't built for the search demands of AI agents. They...
The chDB Durable Layer keeps agent memory fast to query locally while making its analytical state recoverable across laptops, CI jobs, and short-lived sandboxes.
See why today’s AI assistants take minutes to finish simple tasks—and how parallel execution, reusable skills, and fast inference can make them 40× faster.
Google Cloud API Gateway now acts as a native remote Model Context Protocol (MCP) server, eliminating the need to build and maintain custom middleware to expose REST APIs to AI agents. By simply adding specific annotations (like…
Build a multi-account architecture that keeps each team's data in its own AWS account while giving AI agents a unified way to query across them. A central platform account runs the agent using Amazon Bedrock AgentCore Gateway and MCP,…
Agents built with TanStack AI can now call OAuth-protected MCP servers through Vercel Connect, with no credentials for you to store or rotate. The new @vercel/connect/tanstack-ai subpath exports connectMCPTransport , which takes a…
HEMA, a 100-year-old Dutch retailer, turned developer portal-hopping into instant answers by building HAL, an internal AI assistant on Amazon Bedrock AgentCore. Using Model Context Protocol (MCP), HAL delivers governed knowledge inside…
Marking two years of OpenAI Academy and bringing AI skills to even more communities.
GPT-6 Astra produces more structured, context-aware legal documents, freeing lawyers to focus on strategy.
OpenAI and Grab launch GO Forward with AI, a regional programme helping 30,000 partners build practical AI skills across Southeast Asia.
Public sector agencies process large volumes of unstructured evidence, such as body camera footage and scanned documents. This post shows how to combine Amazon Bedrock Data Automation with the Model Context Protocol (MCP) to turn that…
xAI's Grok 4.6 is now available in Amazon Bedrock: a frontier model for long-running agents, coding, and knowledge work, with a 500K token context window and four reasoning effort levels. It runs on both the bedrock-mantle and…
Explore new OpenAI Academy learning paths for employees, developers, leaders, educators, and students to build and demonstrate practical AI skills.
Grok 4.7 from SpaceXAI is now available on AI Gateway as spacexai/grok-4.7 . Requests using this model ID are automatically 40% off through September 27. Grok 4.7 is designed for complex, long-running coding, document, and presentation…
MiMo V2.6 Pro , MiMo V2.6 Flash , and MiMo V2.6 Pro UltraSpeed from Xiaomi are now available on AI Gateway . All three models accept text, image, audio, and video inputs. They support reasoning, tool calling, JSON output, and prompt…
My comment on MCP was always a bad idea? — Hacker News. This article entirely misses the value that MCP brings today. Sure, there's almost no reason to use MCPs if you are running a full-blown terminal agent (Claude Code, Codex, Meta…
mcp-handler now has experimental support for WebMCP , the proposed web standard for exposing tools to in-browser agents. Add a single script tag to your site, and your existing MCP tools become available there too. Opt tools in by…
Kimi K3 from Moonshot AI is now available on Amazon Bedrock, giving you a powerful new open-weight option for coding and knowledge work. It offers native vision, a 1-million-token context window, and explicit prompt caching to reduce…
We dive into these questions and other AI hot takes on the latest episode of the GitHub Podcast. The post Should you read the code, is RAG dead, and did Skills kill MCP? appeared first on The GitHub Blog .
Self-generated prompt injections in compaction summaries In Our framework for reporting model misalignment OpenAI provide "six reports on unexpected or concerning model behavior we’ve observed in the last six months". This one here is…
Google has partnered with Speakeasy to open-source their OpenAPI code generation suite under the AGPLv3 license, a strategic move prompted by the sudden shutdown of Google's previous proprietary SDK provider. The newly open-sourced…
skills@1.7.0 adds Notion skills databases as an install source for agent skills . Notion skills are reusable agent skills written as Notion pages. Teams author, review, and update them in the workspace they already use, then install…
Choosing the right vector store for your Amazon Bedrock Knowledge Bases RAG application affects performance and cost. This post compares Amazon OpenSearch Service, Amazon Aurora PostgreSQL with pgvector, and Amazon S3 Vectors across…
Learn how to enforce defense-in-depth authorization for Model Context Protocol (MCP) tools on Amazon Quick. This walkthrough wires Microsoft Entra ID group and claims-based JWTs through an Amazon Bedrock AgentCore Gateway interceptor to…
AI agents on foundation models often misapply healthcare and life sciences decision frameworks, citing the right guideline but applying it incorrectly. This post shares 38 open-source agent skills across 11 HCLS domains that close this…
Hosted tool execution on the inference backend augments model capabilities and lowers latency.
OpenAI and AARP are bringing free, hands-on ChatGPT workshops to 1,000 older adults across 10 U.S. cities to build practical AI skills safely.
We're working toward bringing parity across Vercel integration and v0 , starting with Resend , Amazon OpenSearch , MongoDB Atlas , Algolia and Clerk . Prompt v0 with what you want to build, and when your prompt requires a provider, v0…
eve agents can now retain context across sessions and use it in future conversations. Persistent memory is organized into slots. You can define named slots in files under agent/memory/ . Each slot specifies a provider, which stores and…
DeepSeek V4.1 Flash is now available on AI Gateway with native image understanding. V4.1 Flash has vision support and accepts text and images in the same request, so you can ask questions about screenshots, read charts, and extract…
When building consumer-facing generative AI applications, balancing high generation quality with fast response times across diverse media types, can be challenging. KDDI, a major telecommunications carrier in Japan, tackled this…
Enterprise work spans messages, decisions, projects, and obligations that change over time. An AI agent that starts without this context must reconstruct it...
new SOTA computer use and coding, 2.5x pricier per token, but WAY cheaper per task, less monitorable. overall, a very successful launch of OpenAI’s new frontier model class.
6 older posts in this topic are not shown; the page lists the 100 most recent of each topic. Narrow the window to bring them within the cap.
Containment, permissions, and the attacks that target agents.
OpenAI and Poland’s NASK are partnering on AI safety, security, and technical research to strengthen safeguards and support a more resilient digital ecosystem.
Samartha Chandrashekar & Nick Parker You click play on a Friday night, and in the second that follows, dozens of services answer in milliseconds. Most of them run on Titus, our internal container platform. Titus is, in essence, a…
Elastic GPUs, sandbox growth, and the economics of the agent cloud
Elastic GPUs, sandbox growth, and the economics of the agent cloud
Samartha Chandrashekar & Nick Parker You click play on a Friday night, and in the second that follows, dozens of services answer in milliseconds. Most of them run on Titus, our internal container platform. Titus is, in essence, a…
Today's PHP Security Release, and How We're Strengthening PHP Security If you're running anything on PHP it's vital you upgrade - today's updates (across the entire PHP 8.x line) address two remote code execution holes, and there have…
Cloudflare has deployed default web application firewall (WAF) protections for newly disclosed PHP vulnerabilities before public disclosure. The rules use request and response signals to stop exploit attempts and prevent exposed data…
Learn how OpenAI built ChatGPT’s LLM-powered ad stack to understand conversational intent, keep ads separate from answers, and deliver relevant ads at scale.
Vercel Drives is now generally available for Vercel Sandbox on Hobby, Pro, and Enterprise. A Drive is persistent storage that you mount as a directory in a Vercel Sandbox. Its files remain available after the sandbox stops, so later…
In August, we launched the $1M hacker challenge for Vercel Sandbox , inviting security researchers to try and escape Vercel Sandbox. We also coordinated with cybersecurity experts Trail of Bits to review the source code and security…
Block is building AI-powered security with OpenAI. Its security team investigates threats faster and proposes patches in minutes.
Baseten partners with Goodfire AI to bring safety controls to open model inference.
Discover how Sophos uses OpenAI’s Daybreak to cut cyber-threat investigation time by 96% and automate 52% of MDR cases while preserving human oversight.
A reference architecture for securely sharing one Amazon SageMaker HyperPod EKS cluster across multiple teams, using AWS IAM Identity Center for authentication, per-team SageMaker Domains and Kubernetes namespaces for isolation,…
A new Vercel Sandbox is now ready to run commands and read or write files when creation completes. Sandbox.create() now resolves only after snapshot restoration completes, so the first command or file operation no longer waits for it.…
Sidecars are isolated containers that run alongside your main Sandbox on the same host.
Vercel Agent can install private dependencies from npm and custom registries using credentials stored as shared environment variables on Vercel. npm , pnpm , and classic Yarn running in Agent sessions authenticate as they do in Vercel…
How ClickHouse’s Postgres extensions handle the memory safety challenges of combining C and C++, from clean language boundaries to isolated helper processes.
Cloudflare Containers now start 6x faster, let your agent choose each sandbox's image and instance type at runtime, and support filesystem snapshots in public beta, all controlled from a Durable Object.
Vercel Sandbox now supports Secure Compute , connecting sandboxes to a team's dedicated network. Public-internet traffic exits through the network's static IPs, and sandboxes can reach private resources in an AWS VPC through VPC…
Cloudflare introduces an adaptive security framework connecting risk discovery, agent governance, runtime protection, and AI-powered response in a continuous learning loop.
Cloudflare learns the structure of your HTTP requests and identifies deviations. You can add a positive security layer that helps reduce attack surface as AI makes it easier for attackers to generate and vary payloads.
Our early guidelines for safety cases in frontier AI training cover technical safeguards, operational practices, and investigating misalignment incidents
Learn how to operationalize Amazon Textract Custom Queries adapters for production: infrastructure as code with AWS CloudFormation and Terraform, a cross-account adapter promotion process, a pre-classification routing pattern for…
External security researchers at Accomplish identified a vulnerability in Cloudflare Containers that could expose residual disk data from previous workloads. We explain how the issue worked, how we investigated it, and the steps we took…
In 2022, we launched a private bug bounty program through HackerOne. Over the last several years, we've worked with HackerOne's VIP program to refine our process, targets, and scope, onboarding thousands of their best researchers and…
Radical Numerics is using biological chain-of-thought and multimodal perception to keep up with the bio-defense arms race, design new genomes and gain insights into biology itself.
OpenAI CEO Sam Altman discusses AI safety, human control, and international cooperation in remarks to the United Nations Security Council.
In August 2026, Hacktron reported what looked like a remote code execution (RCE) vulnerability in Next.js image optimization. Their investigation found that the vulnerable code was not in Next.js itself, but upstream in libheif, an AVIF…
You can now run Harbor evals on Vercel Sandbox. Harbor is the open-source harness behind Terminal-Bench , whose registry includes many other benchmarks such as SWE-bench, tau3-bench and OSWorld. Pass --env vercel to harbor run and each…
Learn how to build a synthetic data augmentation pipeline on Amazon SageMaker AI and Amazon Rekognition that generates photo-realistic, auto-labeled training images for industrial safety AI. This approach improved person detection by up…
A modern storefront can look healthy while malicious JavaScript quietly siphons revenue, hijacks clicks, or rewrites analytics. See how Cloudflare's machine learning models surface evasive client-side attacks for analyst investigation.
We should not treat models as though they have feelings, preferences, rights, or any entitlement to our welfare. Consciousness is the foundation of our ethical, legal, and political systems. To invite another entity to share any flavor…
Vercel Sandbox can now run in all 20 Vercel compute regions , up from four. Running sandboxes closer to the databases, storage, and other services they access reduces latency. Teams can also keep sandbox workloads in approved regions to…
Learn how Heurist built Heurist Finance, a conversational AI investment workbench, on Amazon Bedrock AgentCore. This customer story shows how AgentCore payments, Identity, Memory, Code Interpreter, and Observability let a small team buy…
Amazon Quick custom permissions let you enforce least-privilege access by toggling features per user. This post walks through four patterns to automate custom permissions across the user lifecycle: a RegisterUser API parameter, account…
Requests to Vercel Sandbox public domains are now routed 18x faster. Domains created with the sandbox.domain() SDK call are now resolved from the nearest regional replica, instead of a single centralized store. Incoming requests reach…
Use production traffic and security signals to prioritize findings, prepare edge mitigations when safe, and propose code patches. By combining WAF data with OpenAI Daybreak models, Vulnerability Discovery and Remediation helps teams…
Cursor Cloud Agents can now run in Vercel Sandbox instead of Cursor's hosted machines. Cursor manages the agent harness and inference loop. Its Self-Hosted Machines APIs let you supply the execution environment where agents clone…
Measuring whether an agent works, and keeping it working.
Shipsidekick moved from Neon to ClickHouse Managed Postgres in a single weekend, cutting CPU usage from 100% to 5-10% and making queries about 10x faster.
OpenAI explains how we strengthened containment for frontier model training and evaluation, including offline environments and high-compute red-teaming.
vLLM serves GLM-5.3 on 16 H200s with PCP and DCP prefill and Hybrid HiSparse decode, reaching 1M-token contexts and 1.7–3.2× the per-user speed of data-parallel P/D on the AgentX agentic benchmark.
Learn what impacts OCR accuracy, how it’s measured, and how to improve performance in real-world document processing workflows.
Discover how OCR for invoices streamlines finance operations, automates data extraction, reduces errors, and speeds up accounts payable with LlamaParse.
We added GPT-6 Astra, Luna and Sol, Claude Opus 5.5 and Claude Sonnet 5.5 to the Agentic Analytics Benchmark: all five land mid-board within a point of each other, and Sonnet 5.5 brings the biggest change in turns, cost and truncated runs.
Autonomous AI agents fail softly: latency stays flat while answer quality quietly erodes. This post shares a nine-stage evaluation framework, built on Amazon Bedrock AgentCore, that catches silent quality regressions across the full…
The real challenge of building an agentic system isn't the construction, it's the measurement. Evaluation engineering is a critical foundation for building reliable AI systems. Setting up systematic, repeatable tests to measure your…
Every four years, the World Cup becomes the largest planned stress test of our infrastructure. This year it also set an all-time record! There are a few moments throughout the year, like New Years Eve, that really let us see the…
A practical engineering guide and reference architecture for establishing continuous evaluation pipelines for multi-turn AI agents using Gemini Enterprise Agent Platform, Google ADK, and OpenInference standards.
Where to start with agent evals, offline vs online evals, evaluator types, LLM judges, benchmarks with Harbor, multi-turn evals, and keeping costs down.
By Emma Yanyang Kong , JJ Tan , Ishan Gupta , Lars Olds , Claire Campbell , David Fagnan , Ratna Kavuri , Rohan Gosain , Veli Balin , Minsu Jang , Louis Garcia Introduction Any team generating text with an LLM at scale hits the same…
AI agents offer remarkable productivity gains in complex chip design workflows, but their tendency to generate plausible yet incorrect artifacts makes...
We evaluate several models on 100 tasks from the [internal Binary Exploitation benchmark] (selected at random), and find that GLM-5.3 develops full control flow hijacks in 4% of the trials; Claude Mythos Preview did so in 6%. Although…
Claude Sonnet 5.5 New Sonnet model from Anthropic today. They say it "runs 30%+ faster, and costs up to 30% less for most work" - it's priced the same as Sonnet 5 but appears to beat it on every benchmark, and should be cheaper to run…
How do Postgres providers handle a query that exhausts memory? A recursive query benchmark compares query failures and cluster survival across ClickHouse Managed Postgres, Cloud SQL, PlanetScale, and Amazon RDS.
Wayfair uses OpenAI models to improve ecommerce support and product catalog accuracy, automating ticket triage and enhancing millions of product attributes at scale.
LangSmith Engine now includes Red Teaming to proactively detect agent issues and automated agent testing. Learn more about the Engine v2 release.
Use Jev as a judge for LangSmith evals to evaluate agent traces with faster, cheaper structured feedback across production runs, datasets, and regression tests.
OpenAI outlines a path to shared global AI standards, calling for coordinated evaluation, reporting, and governance to improve safety.
Using GPT-5.6, V7 turns scattered company files into context agents can use to complete complex, source-linked work.
You can now call Jev from TypeSafe AI through AI Gateway using an existing TypeSafe client or the HTTP API, in addition to the AI SDK. TypeSafe client: Point an existing TypeSafe client at AI Gateway without changing its evaluation…
MLPerf Inference v6.1 results were released on September 16, 2026. For AMD, this was a highly successful round in which we achieved several leadership scores, expanded benchmark coverage, introduced a new GPU with validated performance,…
This blog shows you how to reproduce AMD submission results for MLPerf Inference v6.1 on AMD Instinct MI355X, MI350X, and MI350P GPUs using self-contained Docker images, publicly available quantized model weights, and a step-by-step…
Learn how Azure AI Speech LLM 2607 improves multilingual and mixed-language recognition, lowers latency, and simplifies domain-specific customization with enhanced phrase lists. The post Announcing Azure AI Speech LLM 2607: Better…
Multi-turn agents fail in ways single-turn evaluation misses: one early mistake corrupts every later turn. This post introduces the Agent Evaluation Metric (AEM), a decomposable, turn-level way to measure agent quality, applied to its…
Zepto's Push for Reliable, Real-Time Customer SupportZepto is one of India's fastest-growing...
Baseten defines the quality-latency Pareto frontier for STT in Coval's benchmarks.
Pathway's Baby Dragon Hatchling (BDH) is a brain-inspired, post-transformer architecture that reasons in latent space instead of emitting chain-of-thought tokens. See how Pathway develops and scales BDH on Amazon SageMaker HyperPod, and…
Pathway's Baby Dragon Hatchling (BDH) is a brain-inspired, post-transformer architecture that reasons in latent space instead of emitting chain-of-thought tokens. See how Pathway develops and scales BDH on Amazon SageMaker HyperPod, and…
Managed MLflow on Amazon SageMaker AI now syncs richer model metadata (training metrics, evaluation results, inference specs, and lineage) into the SageMaker AI Model Registry, with lifecycle stage promotion. Part 1 shows how to govern…
Wire Amazon Bedrock AgentCore Evaluations into a GitHub Actions pipeline: deploy an AI agent and an OAuth-protected MCP server to AgentCore runtime, invoke the agent with test prompts, score the responses, and automatically block pull…
Serving engines, request paths, and end-to-end latency and throughput.
Deployment summaries now show the size of the build cache each deployment saved and how close it is to the limit for its build machine . The same information is in the CLI with vercel inspect on version 65.1.0 or newer, and in the…
By Roberto Perez Alcolea , Java Platform Team Quasar connects a proposed API change with observed references in consumer repositories, giving the producer evidence for a release decision At Netflix, shared Java libraries support…
Naive byte-range caching can waste gigabytes of origin bandwidth on small requests. We rebuilt our cache architecture in Pingora to coordinate partial fetches automatically, cutting R2 p99 latency in half.
The Global Performance Tuner is a feature for automatic performance tuning released in NVIDIA TensorRT 11.1. It provides automatic, accuracy-aware and...
Deploying a trained neural network involves more than translating each operator into a GPU kernel. It is a compilation problem with many legitimate answers. For...
Updating a SageMaker AI endpoint that has Application Auto Scaling registered makes AWS CloudFormation and AWS CDK deployments fail. This post shows how to build a reusable AWS CDK construct with CloudFormation custom resources that…
Learn how Cerebras serves OpenAI’s GPT-5.6 Sol at up to 750 output tokens per second—without shrinking the model or compromising intelligence.
Day-0 vLLM support for MiniMax-M3.1-Flash on NVIDIA Blackwell, NVIDIA Rubin, and AMD MI350-series GPUs, with NVFP4 MSA kernels, NVFP4 MegaMoE, DSpark speculative decoding, AgentX results, and day-0 RL.
With P/D disaggregation and Wide-EP on a GB300 NVL72 rack, vLLM serves DeepSeek-V4.1-Flash on SemiAnalysis AgentX at 954K tokens/s per GPU at a P90 of 196 tokens/s per user, 7.9x our aggregated single-node result.
Fast, focused, low-cost decisions for any application.
Learn how Cerebras maps trillion-parameter frontier models across WSE-3T wafers to deliver scalable inference at 1,000–2,000+ tokens per second.
At Hot Chips 2026, Cerebras details CS-4 and Nexus, plus the CS-5 and CS-6 roadmap for faster, more efficient frontier AI inference.
See how Cerebras powers GPT-5.6 Sol in OpenAI's Ultrafast mode at up to 750 output tokens per second for time-sensitive AI and agent workflows.
Discover how faster AI inference helps cybersecurity teams improve threat detection, AI security, validation, and response without sacrificing speed.
Cerebras and Upstage bring ultra-fast AI inference to Korea, delivering up to 2,000 tokens per second for real-time enterprise AI applications.
vLLM supports both Flint checkpoints (40B total, 2.5B active) from day 0, decoding a single request at about 500 tokens/s on one GPU.
Run Gemma 4 31B at over 1,800 tokens per second on Cerebras. Build responsive vision, document, screenshot, and agentic AI workflows.
Canadian organizations can now access OpenAI GPT-5.6, GPT-6, and GPT-6.1 models on Amazon Bedrock from the Canada (Central) and Canada West (Calgary) Regions using cross-Region inference. This post covers the models and inference…
This post announces native PyTorch backends for vLLM and SGLang powered by TorchTPU, enabling developers to run and optimize serving workloads using familiar PyTorch tools. Built in collaboration with open-source project maintainers,…
llm-d v1.0 graduates the configuration API to v1 and adds multi-tenant quality of service, zero-downtime operations, and agent-aware serving, taking state-of-the-art distributed inference from a stack that performs to one you can run as…
AI inference is no longer just a throughput problem. As reasoning models, AI agents and interactive applications become mainstream, responsiveness has become a first-class metric. Users notice seconds, not tokens per second. Lowering…
vllm-torchtpu is a new vLLM backend for Google Cloud TPUs built on TorchTPU, with Pallas kernels, multi-host parallelism, KV cache offloading, and disaggregated serving.
Meet the LiteRT Tensor API: giving developers native C++ control over the model graph to unify end-to-end ML computations on heterogeneous edge hardware.
vLLM's prefill-decode disaggregation splits LLM inference across engine instances but must transfer the KV cache between them. This post introduces a push-based NIXL connector that moves transfer coordination off the critical path,…
Multi-turn agentic workloads reprocess growing conversation context each turn, driving up latency. See how bi-directional KV cache transfer lets prefill and decode nodes reuse previously computed context across turns, cutting time to…
A new and tighter integration between Feast and ScyllaDB online feature store, including vector search support and better performance
Mercury 3 from Inception is now available on AI Gateway with 50% off launch pricing. Inception reports strong results across a range of benchmarks, with generation speeds above 1,000 tokens per second. Mercury 3 uses diffusion to…
How asynchronous preparation and NIXL EP enabled near-zero-downtime Elastic EP scaling in vLLM.
Inside the Kimi K3 optimization effort: understanding the bottlenecks and the decisions that made inference faster
vLLM supports Reflection's Beam-1.0 (501B total, 23B active) from day 0, with agent-driven optimization that cut decode latency 2.71× and DFlash2 speculative decoding with adaptive verification.
How agent-assisted CuTe DSL optimizations accelerate Kimi K3 decode on NVIDIA GB300, and how scoped changes, correctness tests, and reproducible benchmarks help ship them responsibly.
Run NVIDIA Nemotron 3.5 Super VL on vLLM for multimodal agent workflows with image, document, and video understanding, OpenAI-compatible serving, and multi-token prediction.
Day-0 vLLM support for Muse Spark Open: 816B MoE, 1M context, multimodal inputs, DSpark speculative decoding, NVIDIA and AMD serving, initial TPU support, and LoRA RL.
Always-on AI agents must do more than generate text. They need to interpret visual information, reason across long-running tasks, and use tools reliably across documents, applications, and video. NVI...
- Model: Muse Spark Open is an open-weight multimodal mixture-of-experts model. It uses an 816B-parameter text transformer backbone, with approximately 42 billion active parameters per token. It accep...
How asynchronous preparation and NIXL EP enabled near-zero-downtime Elastic EP scaling in vLLM.
Learn how to run OpenAI GPT-6 Sol, Luna, and Astra on Amazon Bedrock through global cross-Region inference from Singapore, Jakarta, Thailand, Malaysia, and Taipei. This post covers the Converse and OpenAI-compatible APIs, IAM and SCP…
At Meta, we’ve built a system that enhances ads recommendations by leveraging idle inference capacity in our fleet. Two elasticity levers in the system allow ads recommendation work to expand across opportunistic capacity, without…
Learn how to deploy one large language model across two GPU nodes on Amazon EKS using the multi-node Ray LLM Deep Learning Containers, with separate prefill and decode nodes behind one OpenAI-compatible endpoint. The pre-built container…
Test and promote inference endpoints to evaluate quality, latency, throughput, and cost.
Bi-directional KV transfer lets vLLM prefill nodes reuse previously computed KV from decode nodes on later conversation turns, cutting redundant prefill recompute and reducing TTFT by up to 3.3x on long multi-turn prompts.
New Pro teams now default to a 30-day deployment retention period. Coding agents have accelerated development, increasing deployment frequency and the amount of stored deployment history. Shorter retention keeps recent history available…
Deployment Storage and Functions Storage billing will begin for all Pro teams at $0.10 per GB-month. Check your email for details about when billing will start for your team. To keep storage costs low, the deployment retention window…
vLLM now runs on NVIDIA Vera Rubin NVL72 with day-0 model support, Rubin-tuned FlashInfer kernels, and locality-aware MoE, delivering 7.8x the per-GPU throughput of GB200 NVL72 on AgentX.
We had early access to two NVIDIA Vera Rubin nodes (8 GPUs) and used them to bring SGLang and Miles up on the platform: SGLang for inference and Miles for RL training. On the inference side, we tuned ...
OpenAI’s Decisions API is now available through AI Gateway at /v1/decisions . It uses the same request and response format as OpenAI, so existing OpenAI SDK integrations can switch by changing the base URL, credentials, and model ID.…
Within three weeks of release, vLLM made DeepSeek-V4.1-Flash 1.9x faster at low concurrency and lifted its throughput 5x on SemiAnalysis AgentX, with SWA bounded replay, CUDA graphs, DeepSeek's new kernels, and vLLM kernel fusions.
Claude Code with Fable 5 built a Qwen 3.6 inference engine that beat vLLM by up to 90% on decode speed and 2.3x on TTFT, using MetaInfer skills on one B200.
We achieved 200+ tokens per second on GLM-5.
Cloudflare AI Gateway now supports native web search API integration in partnership with Ceramic.ai, Exa, and Linkup. Developers can now inject real-time web context into model inference calls via AI Gateway, REST APIs, or Workers bindings.
Amazon Bedrock now supports Anthropic's Claude Opus 5 and Claude Sonnet 5 with in-region inference in Seoul, and Claude Sonnet 5 in Singapore. If you have local data processing requirements in South Korea or Singapore, you can now use…
Parallel work, model-family isolation, reversible changes, and GPU-backed validation shaped an open source project designed around coding agents NVIDIA TensorRT...
Deploy a text-to-speech model on Amazon SageMaker AI with the AWS vLLM-Omni Deep Learning Container and stream generated speech over a persistent bidirectional connection. This Part 1 tutorial deploys Qwen3-TTS and streams speech…
Deploy two generative media models from one AWS vLLM-Omni Deep Learning Container on Amazon SageMaker AI. Generate an image with FLUX.2-klein through real-time inference, then animate it into video with Wan2.1-VACE through asynchronous…
A customer asks whether an order has shipped. An agent already has the order ID and three possible next actions: query the order-status service, search general delivery-policy documentation, or ask th...
Learn how faster AI inference helps cybersecurity teams analyze threats, accelerate triage and response, and contain attacks before they spread.
How vLLM implements distribution-preserving Gumbel-max text watermarking with efficient GPU kernels, statistical detection, speculative decoding, and repeated-context safeguards.
Maximizing perf on AI-SQL queries with the KV-optimal left-deep join
An AI coding agent’s patch can pass tests yet fail when the server loads a real model and handles requests. Evaluating changes to inference-serving software...
Moonshot AI released the weights for Kimi-K3, a 2.8-trillion-parameter, 1M-context, natively-MXFP4 Mixture-of-Experts (MoE) model, and AMD Instinct™ support was available on day 0 across three independent serving frameworks: vLLM,…
vllm-metal brings vLLM's paged, continuously batched serving stack to Apple Silicon, with flatter TTFT under concurrent agent load, batched MTP, and automatic M5 prefill acceleration.
Shipping Python bytecode saved about 18 seconds per startup in H100 server benchmarks, with images and models already local.
The compute and memory demands of generative AI increasingly exceed what a single GPU can provide. NVIDIA TensorRT multi-device inference is a new capability...
Learn about the newly launched version of the Dropbox API documentation and its modern design and features.
How to leverage NVIDIA Hardware Video Decoders to Achieve Multi-GPU Scaling in Video Captioning and Description tasks.
GLM 5.3 FlashX is now available on AI Gateway. GLM 5.3 FlashX is a high-speed serving option for Z.ai's multimodal coding model, delivering inference at ~200 tokens per second for faster streamed responses. The higher serving speed is…
You can now opt into Turbo build machines on any individual deployment. This is useful when you need to increase resources temporarily without changing project settings. You can do this in three ways: Include #VERCEL_BUILD_MACHINE=TURBO…
Amazon SageMaker HyperPod now supports model caching for inference, which pre-loads model weights and container images onto cluster nodes so pods read from local NVMe storage instead of downloading over the network. Learn how model…
Biomolecular structure prediction is now often run at proteome scale, where the goal is to move an entire worklist through the pipeline efficiently. NVIDIA...
A performance model for LLM serving: inspect local shapes, remove repeated work, verify data movement and dispatch, then follow the queue.
Every request to Vercel passes through our CDN , which executes on average over 80 million routing instructions per second. Part of that work is looking up metadata to determine which paths exist and how to serve them. When that…
GPT‑Live‑1 brings natural, full-duplex voice conversations to the API, with stronger instruction following, custom voices, and telephony support.
DeepSeek-V4.1 introduces several architecture choices that shape the serving stack. Low-ratio compression and sliding-window attention (SWA). Each layer maintains an fp8 sliding-window cache for its ...
Learn how to deploy Qwen3.8-2.4T-A95B, a 2.4-trillion-parameter open-weight model, on Amazon SageMaker HyperPod with vLLM. This walkthrough covers cluster provisioning, NVFP4 quantization, and an OpenAI-compatible endpoint with built-in…
TorchServe is no longer maintained, leaving teams to own the entire GPU inference stack. The AWS Ray Serve Deep Learning Container is a supported, pre-tested container with the framework, GPU drivers, and serving layer already…
vLLM integrates HiSparse as a pressure-driven memory tier that composes with the Hybrid Memory Allocator and KV offloading, letting GLM 5.3 requests keep decoding when their KV no longer fits in GPU memory, so concurrency stays high.
How vLLM optimizes KV cache management, parallelism, scheduling, and P/D disaggregation for agentic workloads, validated on SemiAnalysis AgentX with up to 130K tokens per GPU-second and a 14.6x-106x serving-cost advantage over Opus 5.
Tenstorrent accelerators join vLLM as an out-of-tree platform plugin, driven by mesh-architecture choices: phase-based scheduling, single-process data parallelism on Galaxy, on-device sampling with host fallback, and async decode overlap.
vLLM integrates HiSparse as a pressure-driven memory tier that composes with the Hybrid Memory Allocator and KV offloading, letting GLM 5.3 requests keep decoding when their KV no longer fits in GPU memory, so concurrency stays high.
Reusing computed attention state across requests and turns.
The KV cache is a fundamental building block of the modern LLM inference system. The context from multiple conversation rounds in agent sessions is cached as keys and values (KV) in GPU memory, allowi...
Prompt caching in Amazon Bedrock can cut input token costs by up to 90% when you repeatedly send the same context to foundation models. This post walks through six practical prompt caching scenarios using the Converse API: message…
How requests are batched, placed, and split across a fleet.
Teams deploying custom text-to-speech models on Amazon SageMaker AI often guess at autoscaling thresholds, and a wrong guess either wastes GPU spend or breaches a latency SLO. This post deploys Kokoro-82M as a custom container and…
Learn how AI document classification automates document sorting and routing at scale, and what separates systems that perform on real documents from those that don
Voice agents, multiplayer games, and other workloads can now build on Sticky Sessions. Achieve strong affinity without losing autoscaling. Here’s how we built this on Modal Servers.
Smarter routing for subnet routers, app connectors, exit nodes, and Services.
Enterprise teams can now run Vercel Functions in their own AWS account with Bring Your Own Cloud (BYOC). With BYOC, your function code and execution stay inside your security and compliance boundary. Vercel continues to handle…
Vercel Connect now supports Secure Compute , allowing connectors to reach providers with IP allowlists, such as Salesforce and Snowflake. With Secure Compute routing enabled, Connect’s provider requests use the selected network’s…
d1-ultra from Liquid AI is now available on AI Gateway. It is a multimodal decision model for classification, routing, and scoring across text and images, including visual defect inspection and checking damage claims against customer…
This post is part of the AMD Disaggregated Inference (DI) Blogs Series, which takes frontier open models from a single-node demo to production, multi-node serving on AMD Instinct™ GPUs
Liquid d1 is now available on AI Gateway . d1 is a decision model that evaluates shared state against typed questions for classification, routing, and scoring. It returns structured answers with probabilities without generating text…
Whether prefill context parallelism pays depends on prompt length. It divides the prompt across GPUs during prefill; on four AMD Instinct™ MI355X GPUs (gfx950) serving GLM-5.2-MXFP4 under SGLang at --tp-size 4, enabling it yields 43% to…
An oracle router over 18 models scores 97.6% on DeepSWE at $1.88 a task, vs. 74.1% for the best single model. See how FireRouter picks per task.
Amazon SageMaker AI shipped 13 inference launches in year-to-date across two deployment paths: fully managed endpoints and Amazon SageMaker HyperPod Inference. This post reviews each launch, from inference recommendations and…
Amazon SageMaker HyperPod Inference Gateway is a Kubernetes-native, GPU-aware routing add-on for Amazon EKS. It uses real-time GPU signals to send each inference request to the best-suited pod, cutting first-token latency by up to 82%…
Jev from TypeSafe AI is now available on AI Gateway . Jev is a probabilistic decision model for software: state goes in, typed Choice, Score, and Boolean answers come out. Regular language models generate text one token at a time, which…
Temporary tables and named sessions live on a single ClickHouse replica, so a follow-up query routed elsewhere can't see them. Replica-aware routing pins your requests to the same replica over HTTP or the native protocol — and here's…
Discover DeepSeek-V4.1-Flash, a 552B multimodal MoE model featuring a Causal Encoder-Decoder architecture and efficient KV caching.
Tailscale on Vercel Hundreds of AI models shipped to customers in-product Model access granted and revoked by tailnet network identity Went from model routing prototype to paying customers in months Tailscale connects a company's…
Amazon SageMaker Inference now offers prefix-aware routing, a routing strategy that sends requests sharing the same prompt prefix to the same instance so the KV cache stays warm. In benchmarks on Llama 3.1 70B, it reduced P50…
The disaggregated storage model of Lakebase Postgres provides a feature rich, flexible...
The shapes of the models being served and how tokens come out of them.
Glimpse from Elorian is now available on AI Gateway. The visual reasoning model accepts images and video to follow objects through motion, interpret spatial relationships, and connect an action to its physical consequences. Glimpse can…
Nemotron 3.5 Super VL is now on Baseten. It is an open 120B-A12B MoE model built for agent workflows across text, images, and video.
Jev 1.14 from TypeSafe AI is now available on AI Gateway . Jev is a decision model for software. You pass it state and a set of questions, and it returns typed Choice, Score, and Boolean answers with probabilities instead of free-form…
Nemotron 3.5 Lightning from NVIDIA is now available for free on AI Gateway as nvidia/nemotron-3.5-lightning-free , served by NVIDIA. Nemotron 3.5 Lightning is an open-weight reasoning and chat model for long-running agents, sub-agent…
An online-fitted acceptance estimator that lets every draft-model speculator use Adaptive Verification, by predicting the target's acceptance probability from the drafter's own logits and refitting continuously during serving.
Serving a frontier Mixture-of-Experts model like GLM-5.3-FP8 at long context and high concurrency is a demanding workload. The model has 744 billion total parameters, with approximately 39 billion active per forward pass, and a native…
Gemini 4 Argon is a general-purpose physical reasoning model designed to enhance workflows across the robotics stack with advanced capabilities in physical understanding, code generation, and visual-semantic reasoning. Developers can…
AWS engineers share kernel optimizations for fused Mixture-of-Experts (MoE) models on Amazon SageMaker AI: Swap-AB for WGMMA utilization, Triton backend routing for FP8 quantization, and eliminating Triton JIT recompilation. Together…
Autoregressive decode generates one token per model forward pass, so single-request latency is bounded by how fast you can stream weights through the GPU — not by compute. Speculative decoding attacks that bottleneck by letting a small…
How can a 30B-parameter model activate only 3B parameters per token, and still use the capacity of the larger model? Nemotron 3.5 Lightning illustrates the...
Loops provides managed infrastructure for fine-tuning models through an API, with support for SFT, RL, and long-context workloads.
DeepSeek-AI released the DeepSeek-V4 series on April 24, 2026: a preview pair of MIT-licensed Mixture-of-Experts models, with DeepSeek-V4-Flash at 284B total parameters (13B activated) and a one-million-token context window. Flash…
From loop engineering to harnesses, squads, and open weights, the GitHub Podcast breaks down the AI terms showing up in developer conversations. The post Decoding the new AI lingo: Loops, harnesses, squads, hill climbing… oh my!…
This post is the third in a series on AI model co-design. It explores how to accelerate LLM inference while maintaining accuracy using speculative decoding and...
The arithmetic itself: kernels, numerics, and compilers.
Introducing Humming, a quantized inference kernel library from Ant Group and the vLLM community, with broad GPU and quantization support, dense and MoE performance results, and vLLM integration.
Google AI Edge Model Explorer now supports custom compiler intermediate representations and open-source adapters like TOSA, VGF, and PTE for on-device machine learning. Developers can leverage hierarchical graph visualization,…
Once you have identified a GPU kernel to optimize, you need to understand how its instructions execute and where waves spend time waiting. Thread trace gives you a detailed view of that behavior, helping you investigate stalls, gaps in…
Novita AI has open-sourced Chord, a high-performance W4A16 MoE CUDA kernel for Kimi K2.x serving shapes, with a Humming-compatible indexed path and grouped SM90 operators.
We ported ThunderKittens to NVIDIA's Vera Rubin NVL72 and rebuilt our NVFP4 GEMM around the new hardware, taking it from 42% of roofline to over 22 PFLOPS — competitive with cuBLAS and CuTe DSL. Here is what changed in the ISA and how…
Using quality-aware quantization, index-based clustering, and scheduling optimizations
A TensileLite run can only return a kernel its configuration’s search space can express. That sounds like a technicality, but it is the one fact that decides whether a tuning campaign helps or hurts. The configuration is not a hint…
In September 2026, NVIDIA announced it is leaning into native GPU programming in Rust. CUDA C++ and CUDA Python are mature, enterprise-grade toolchains, and...
Silicon, interconnect, and the fabric between them.
Meet Cerebras CS-4, a wafer-scale AI accelerator with up to 30× faster inference than GPU systems, native disaggregated inference and higher throughput.
World Action Models (WAMs) offer a promising route to more physically grounded robot policies, but their video-model backbones make inference expensive. In this blog you will learn how to remove much of that cost on AMD GPUs without…
TorchTPU integrates a native torch.device("tpu") backend directly into PyTorch, allowing developers to train large-scale models on Google Cloud TPUs without rewriting code in JAX. The TorchTitan TPU fork enhances this with…
Google Research has completed the 32-bit open-source Coral NPU RISC-V architecture with the introduction of its Matrix Execution Unit, which delivers up to 1 TOPS while targeting 10 mW average power for edge inference. Alongside this…
Every whole-slide image tells a story in pixels. A whole-slide image is a glass pathology slide scanned into a single very large digital image, stored as a pyramid of small tiles so a viewer can pan and zoom it. A single biopsy…
rocOpt is a new open-source, GPU-accelerated engine for decision optimization on AMD Instinct GPUs. It solves large-scale linear programming (LP), mixed-integer linear programming (MILP), quadratic programming (QP), and vehicle routing…
Nvidia’s Vera server chip aims for high per-core performance.
Editor's note: Generative AI startups are moving rapidly beyond static clip generation to build interactive, real-time worlds. In this technical deep dive, we explore how AI startup Visko collaborated with Google Cloud TPU engineers to…
How to run a LiveKit video avatar with open-weights models on Modal GPUs, using Sticky Sessions for container affinity.
Our strategies to optimize our data pipeline throughput without compromising data quality.
We trained a model on agentic coding tasks using fully asynchronous RL and mixed-precision, mixed-accelerator samplers. With off-policy correction, training stayed stable through convergence, on par with homogeneous samplers, at a lower…
Introducing Mistral Large 4: Le chonk Mistral are back in the game. Today they're releasing a preview of Mistral Large 4, a 1 trillion parameter, 49 billion active parameter model trained on their own cluster of 3,800 NVIDIA Grace…
GPU applications increasingly need networking and data movement to behave like first-class GPU-controlled operations rather than host-driven services. When the...
EmbeddingGemma 2 is a new 740M open-weight multimodal model that maps text, images, video, and audio into a unified vector space for privacy-first, on-device retrieval. Developers can easily integrate these capabilities cross-platform…
GPU applications increasingly consist of multiple independent components running at the same time within a single process: a latency-sensitive operator...
This post is a continuation of the Multi-Accelerator Support for AIMs and AMD Solution Blueprints blog, which introduced (i) multi-accelerator coverage across AMD Instinct™ GPUs, Radeon™ PRO GPUs, and EPYC™ CPUs, and (ii) walked through…
Teaching a robot to “put the bowl on the plate and place the cream cheese inside the bowl” used to require a rack of data-center GPUs sitting behind the robot. In this blog we show how a modern vision-language-action (VLA) model,…
ClickHouse expands its Microsoft collaboration with a native Fabric workload, OneLake read and write capabilities, and ClickHouse BYOC availability in the Microsoft Marketplace.
ClickHouse Workload for Microsoft Fabric brings sub-second analytics to OneLake data through a native Fabric workload, with dedicated ClickHouse Cloud compute for agents, dashboards, and applications.
Every NVIDIA CUDA Toolkit release adds functionality and performance improvements that help developers get more from NVIDIA GPUs and the broader NVIDIA software...
The other half of the fleet: pre-training, post-training, and RL loops.
Multi-Turn Reinforcement Learning (MTRL) on Amazon SageMaker AI trains models on the outcome of a whole multi-step task rather than labeled steps. This hands-on guide walks the full pipeline using Wordle: deploying the agent and reward…
Access a network of highly vetted partners to build custom RL environments to accelerate your post-training efforts.
Amazon SageMaker AI serverless model customization now includes training sessions, generally available today. You write and coordinate your own post-training loop (custom loss functions, reward signals, or multi-turn reinforcement…
Amazon SageMaker AI training session APIs let you write your own LLM training loop (SFT, RL, or custom objectives) in a Python script while the service runs the GPUs. This post introduces the programming model and three SDK clients,…
Spindle is an open-source API for training and fine-tuning open models — deploy a server with a single command and run any Tinker-compatible training code against it.
Reinforcement learning for LLMs is hard, so we built Modal Dojo — a library that gives you the building blocks, tuned defaults, and observability to get started with production-grade LLM training.
What we shipped at Interrupt NYC and London: LangSmith AI Search, Engine v2, Trajectories, Custom Apps, Fine-Tuning, and Managed Deep Agents v0.9.
In this blog you will learn how to pre-train hybrid LLMs from scratch with Primus on AMD Instinct™ MI355X GPUs. In particular, we will cover how to declare a hybrid layer stack with a single configuration string, how three…
We are introducing Clef and Clef-flash, open-source decision models hosted on Workers AI for high-speed classification and agentic workflows. Also launching: a new reinforcement learning platform that allows developers to fine-tune…
Automatic speech recognition must handle how people actually speak, not only the languages and styles that dominate pretraining data. Regional dialects and...
Learn how to scale Mixture-of-Experts (MoE) reinforcement learning on Amazon EKS using Elastic Fabric Adapter (EFA) and DeepEP. This post presents an architecture that combines Amazon EKS, EFA, and Amazon S3 and increased aggregate…
Learn how to run SkyRL, an open-source reinforcement learning framework, on Amazon SageMaker HyperPod to post-train a Qwen3-VL-8B vision-language model with GRPO. This walkthrough covers building the container image, launching a Ray…
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 like Gemini. So here at Google…
Amazon SageMaker HyperPod and Cloud Native Qumulo let you place training compute in one AWS Region while keeping your dataset in another. This post shares the architecture and validation results from a cross-Region training run, where a…
LangChain announced new updates to LangSmith. Updates include Engine v2 with red teaming and automatic testing, a new version of Managed Deep Agents, trajectories and more.
LangSmith Fine-Tuning is in public beta today, and its open-source CLI, smithtune, trains on Baseten Loops.
The MaxText team successfully reproduced Ai2’s Olmo 3 7B language model from scratch on Google Cloud TPUs using JAX/XLA, precisely matching the original PyTorch-on-GPU reference across pre-training and mid-training stages on all…
LangChain introduces LangSmith Fine-Tuning and SmithTune, a CLI built for post-training models. Train specialized models without building data pipelines by hand.
Fireworks’ ARCv3 delivers lossless BF16 weight updates nearly 50% smaller than ARCv2, making cross-region reinforcement learning more practical.
In online reinforcement learning (RL), a language model generates responses, receives rewards, and updates its weights to improve future responses. Many systems split this loop between a rollout engine, which generates the responses,…
Amazon SageMaker AI now offers instance preference lists for training and processing jobs. Specify an ordered list of up to five instance types, and SageMaker AI automatically launches on the first type with available capacity,…
Cloudflare is giving site owners a way to stay discoverable while disallowing AI training. New controls and an Accountable designation establish a shared model with Apple, Google, and Microsoft.
Pick the right generative AI customization approach on AWS with an 8-step decision framework, from prompt engineering and RAG to fine-tuning, continued pre-training, and Amazon Nova Forge. Start simple and escalate only when you must.
Fyxer uses OpenAI models, fine-tuning, memory, and real user feedback to organize inboxes and draft emails in each user’s voice.
Demonstrates llm-d time-slicing in a real-world use case: running more post-training experiments, SFT and full-parameter RL on different base models, concurrently on the same GPUs using the time-slicing stack via OpenRL.
Compare top open-source models for post-training. Learn what drives cost and which model fits use case.
What the fleet costs to buy, run, and power.
The generative AI functions in BigQuery ( AI.GENERATE , AI.IF , AI.CLASSIFY , AI.SCORE , and more) bring large language models (LLMs) directly into your SQL, so a single query can classify millions of support tickets or score each…
Guest Post: In science, thinking has gotten cheap but doing has not. This asymmetry is reshaping how research companies operate, largely inconspicuously.
Following our announcement of Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking last week, we are thrilled to share that Gemini 3.8 Live with Live Avatar is now generally available in Gemini Enterprise . First previewed at Google…
Yesterday was Grok 4.7 ( pelicans ) and MiMo v2.6 Flash/Pro ( more pelicans ). Today Anthropic released Claude Opus 5.5 , and around an hour later OpenAI released GPT-6 Sol and GPT-6 Luna . It's going to take a while to get a good read…
Fountain cut analytics latency from three hours to under two minutes and costs by 66% by rebuilding Cue’s data plane on ClickPipes and ClickHouse Cloud.
GPT‑6 Astra allowed Parallel’s agents to research and synthesize labor-market data in half the time and at half the cost vs. prior models.
Genspark partnered with Fireworks Lab to post-train MiniMax M3 into Gen-1 Slides, a model that produces Opus 5-level decks at roughly one-tenth the cost per finished deck.
Together GPU Clusters now supports preemptible compute: the same GPU capacity at a flat 50% of the on-demand rate, with a five-minute drain window.
Durability, query execution, and the layers under both.
I have unkind things to say about people who make their own pasta. Here, I’ll only use them as an example of how not to think when you’re building infrastructure. It’s unfairly easy to believe that the harder you work on something, the…
We moved a copy of 99.99999% of GitHub's repositories to Azure. Here's how per-reference quorums will bring over the rest, and how object storage lets reads scale independently. The post The bridge to agent-scale Git infrastructure…
Cloudflare's web application firewall can now correlate an HTTP request with its response. Full transaction detection helps every customer find successful exploits, data exposure, and risky misconfigurations that request-only inspection…
NVIDIA Vera tops the ClickBench hardware leaderboard, and three micro-benchmarks for filters, aggregation and joins show ClickHouse scaling almost linearly across its 176 cores.
A deep dive into how D1 architected its control plane to scale and manage millions of databases, while keeping data plane query operations at the edge. We share guiding principles on how to build on the Developer Platform effectively.
Learn how Deep Agents SDK manages context for long-running AI tasks through offloading, summarization, and filesystem abstraction to prevent context rot.
You can now run Redis commands, browse keys, and inspect values directly in the Vercel dashboard using the Marketplace database browser . Both the Redis and Upstash for Redis integrations in Vercel Marketplace are supported, so you no…
Over the weekend, “System One” decision models such as Jev have launched, which are...
Fireworks introduces the Specialized Intelligence Index, a one-stop destination for real-work benchmarks across industries.
Compare open, closed, and specialized models on practitioner-built benchmarks for healthcare, legal, finance, cybersecurity, support, and software.
See how local NVMe transforms Postgres transaction performance—and why ClickHouse remains essential for fast analytics as workloads scale.
TigerBeetle is a transaction-processing database. Its double-entry accounting primitives are simple and powerful: they can represent the movement or exchange of any quantity, including financial transactions such as real-time payments,…
Hobby projects now retain fewer deployments past the 30-day retention window. Hobby teams get 10GB of Deployment Storage . Every deployment you keep uses some of it, and going over the limit can block you from deploying until you free…
Explore ClickHouse AI Functions for classification, generation, translation, embeddings, semantic search, and cost controls—all directly from SQL.
Learn how OpenAI evolved Habitat from a Python library into a globally distributed storage platform serving 1 billion ChatGPT users and 22M requests per second.
Agreement, addressing, and moving bytes between machines.
Learn how to connect on-premises data sources to Amazon Quick privately with AWS PrivateLink, keeping analytics traffic off the public internet. This walkthrough covers VPC endpoints for user and API access, VPC connections for data…
An open source “bomb” drops from China
An open source “bomb” drops from China
We think "did a human read the code?" is an imprecise proxy for "what could go wrong with this change, and which check—automated or human—is the right one to catch it?" Asking that question also changed when review happens. Before,…
ClickHouse’s new native TCP client for .NET brings faster reads and inserts with fewer allocations, columnar APIs, POCO support, and a path to migrate from HTTP.
At the 2026 Open Compute Project Global Summit, Meta is sharing the next generation of open networking infrastructure, including: Our first liquid-cooled network rack (IcePack), and 102.4T switch platform (IceCube). Flattened Planar…
Meta has begun deploying Media over QUIC (MoQ), an emerging open protocol for real-time video streaming at scale. MoQ will help simplify our protocol stack for live streaming video and enhance overall user experience for live video.…
We’re introducing ZGateway, the proxy we are using to unify traffic through ZippyDB, Meta’s most widely-used key value store. As a bonus, it also enables admission control, load balancing, cross-region resilience, and richer operations.…
AI agents are learning to do more by working together. A lead agent can break a complex task into smaller jobs and assign those jobs to specialized subagents....
The unit of deployment and the machinery that starts it.
Muse Spark Open, Meta’s open weights model, is live on Fireworks: #1 on Vals AI’s Harvey Legal Agent Benchmark, on US-only serverless with zero data retention.
Imagining traditional gearheads getting their hands on software-defined vehicles is an amusing thought experiment. With their socket wrenches and air compressors at the ready, they might be confused to encounter core functions more…
Sometimes, giving an AI agent all of the knowledge it needs is the wrong approach. At Meta, we’ve seen notable improvements in efficiency and accuracy by keeping the orchestration layer of our data pipeline agents deliberately thin: It…
At Databricks, we continuously develop and evaluate new AI research techniques to...
Cloudflare Gateway and Data Loss Prevention have inherited the flexibility and power of Cloudflare Workers through our full-stack integration with the Cloudflare Developer platform.
Last year, we expanded zero-config support for Python and TypeScript backend frameworks. By this summer, developers could deploy their frontend and backend together with Vercel Services and coordinate background work with queues and…
Vercel Daemons is now in public beta, allowing you to run long-running processes that pull and execute work from Vercel Queues , databases, or third-party systems without requiring an HTTP server: Consumers using Vercel Queues poll mode…
By Violetta Pidvolotska and Naveen Mareddy Stratum is Netflix’s internal serverless media processing platform, shaped by over a decade of experience developing large-scale distributed systems. It powers a wide range of media workloads —…
AI is mass decompiling games. That's both awesome and scary. Indie game developer Popcar talks about the recent wave of AI-powered decompilation of existing games, enabling weird mashups between existing games, and producing a boom in…
Learn how Tamnoon built Tami, an agentic cloud remediation orchestrator on Amazon Bedrock that autonomously investigates and safely remediates cloud security findings from CNAPP and cloud-native sources. The approach drove a 90%+…
Training or finetuning a multi-billion parameter foundation model requires significant compute...
ClickHouse Cloud delivered 1,996× better performance per dollar than Redshift Serverless in CostBench. We trace the gap from fresh data arriving to fast answers coming back.
Last week we announced that Gemini Enterprise has evolved to make it even simpler to get work done, and introduced the Gemini agent , your new universal agent for work. It's always on and ready to answer your questions. It will write…
Deno is joining Cloudflare The Deno team released the first version of celld back in August - their open source implementation of the Durable Objects pattern from Cloudflare Workers. Today, Cloudflare are acquiring Deno outright, with…
iSAM Funds uses ClickHouse Cloud to research full options history, achieving 35–50x compression and 100x faster single-core ingestion while simplifying its data pipelines.
On-demand CPU and memory profiling for Cloudflare Workers and Durable Objects is now available. Now you can generate interactive flamegraphs directly in production to quickly track down memory leaks and performance bottlenecks.
The Deno team is joining Cloudflare to radically simplify self-hosting Workers and Durable Objects, so developers can use the same primitives in more places.
Connect to ClickHouse Cloud with short-lived tokens from your identity provider instead of database passwords.
ClickHouse Cloud delivered 438× better performance per dollar than BigQuery in CostBench. We trace the gap from fresh data arriving to fast answers coming back.
GPU-accelerated Kubernetes clusters depend on compatible versions across dozens of components, each on its own release cycle: host kernels, GPU drivers,...
ClickHouse Cloud delivered 752× better performance per dollar than Databricks in CostBench. We trace the gap from fresh data arriving to fast answers coming back.
Release: pwasm 0.2a0 pwasm is one of my folly projects - an entirely vibe-coded pure Python WebAssembly engine that I built in January during my first bout of AI mania . I hadn't touched it since January, so I decided to let Claude Opus…
Workers KV Instant delivers sub-2ms p99 read latencies and 250ms global replication across Cloudflare’s 300+ edge locations. KV Instant eliminates cold-read penalties and uses the familiar Workers KV API.
Cloudflare K2 is a serverless event streaming service built directly on top of R2 object storage for high-scale data movement and long-term retention. By decoupling producers and consumers at the edge, K2 enables durable, ordered log…
Cloudflare is hosting a competition to see who will build the next Git platform for an era of AI agents. Artifacts is in open beta, with Workers bindings, data jurisdiction controls, and event subscriptions for repository changes.
Cloudflare Workers is adding opt-in support for post-quantum-resistant algorithms ML-KEM and ML-DSA. You can try it today.
Cloudflare Basin is built on Apache Iceberg and R2 Object Storage. With general availability, developers can ingest, manage, and query large datasets at scale without paying data egress fees.
Everything announced live at Modal's inaugural conference, Runtime.
ClickHouse Cloud now supports SCIM 2.0, so your identity provider can create, update, and deprovision organization members and sync group membership to ClickHouse roles automatically.
Cloudflare Registrar’s new search delivers fast, transparent results across 420+ extensions using Workers, Durable Objects, and WebSockets. Its expanded API and cf CLI also let agents search, register, and transfer domains.
Cloudflare is launching The Cold Start, a startup competition giving five early-stage companies five minutes on stage at Cloudflare Connect. Grand Prize winner receives $500,000 in credits, a San Francisco billboard, and an invitation…
With the new experimental support for the Emscripten target in Rust Workers, many previously unsupported Rust libraries and applications can now be built and deployed directly to Cloudflare’s global Workers platform, including upcoming…
By Dhruv Pratap Introduction Organizations that have been around for a while usually run two identity systems side by side. One belongs to the cloud provider: IAM roles, instance profiles, execution roles. The other is your own, and it…
You can now push container images from GitHub Actions to Vercel Container Registry (VCR) without storing long-lived registry credentials. The new vercel/vcr-action/login action authenticates your workflow using GitHub OIDC. It exchanges…
Cloudflare Python Workers are now generally available After a two year preview, Cloudflare's support for running Python code in their server-side Workers platform is now stable: "Python is now a first-class, fully supported language on…
BMW Group operates CLEA, a FinOps platform monitoring more than 14,000 cloud accounts. This post shows how BMW added automated daily cost anomaly detection, moving from reactive dashboards to proactive alerts using Prophet forecasting,…
Python Workers allow developers to run Python web frameworks and AI orchestration libraries natively in the Cloudflare Workers runtime. You can seamlessly integrate with Cloudflare's ecosystem including D1, R2, and Workers AI without…
Builds using Secure Compute or Static IPs now start 64% faster, with the average time from deployment creation to build start dropping from 6.7 seconds to 2.4 seconds. Previously, each build waited for a new build container to boot with…
Welcome to the first Cloud CISO Perspectives for September 2026. Today, Sandra Joyce shares the latest details on Google’s visibility into how attackers are using AI, and how we’re using AI to stop them. As with all Cloud CISO…
Learn how to automate end-to-end PII detection and redaction from scanned documents at scale using Amazon Bedrock Data Automation with a custom blueprint, AWS Step Functions, and AWS Lambda. A custom blueprint redacts sensitive fields…
New OpenAI Economic Research shows how workers use AI beyond traditional roles and which new activities become recurring parts of their work.
Manually tagging thousands of catalog products is slow and inconsistent. This walkthrough shows how to customize Qwen3-8B with supervised fine-tuning (SFT) and reinforcement learning with verifiable rewards (RLVR) on Amazon SageMaker…
Federated learning (FL) projects often begin with a straightforward setup: one server, a few clients, and one dataset at each site. As those projects grow, the...
You can now scope access to individual Workers and assign narrower Developer Platform roles, so teammates, CI tokens, and agents get only the access they need to debug, deploy, or monitor safely.
Any Nix package, live in your browser Farid Zakaria calls this his " magnum opus of Nix work", and I can see why. trynix.dev provides a qemu-wasm powered x86_64 Linux virtual machine running entirely in your browser through WebAssembly.…
ClickHouse On-Demand Compute lets you scale individual queries with additional workers, run intensive workloads without disrupting production, and use compute when you need it.
ClickHouse Cloud delivered 412× better performance per dollar than Snowflake in CostBench. We trace the gap from fresh data arriving to fast answers coming back.
ClickHouse Managed Postgres is expanding to Google Cloud, bringing NVMe-backed storage, native CDC into ClickHouse, and a unified query layer via pg_clickhouse to GCP private preview customers.
Faster cross-cluster connections, fewer IPv6 dead ends, less certificate churn.
Workers now enables Node.js compatibility by default, supports applications up to 64 mebibytes, and adds a URL-based module registry with import.meta, lazy compilation, shared code caches, and clearer errors.
Modern AI platforms are no longer a single application behind one login screen. A user may start in a central portal, open a governed dataset, launch a notebook...
Learn how to embed individual Amazon Quick Sight visuals into a React application with per-user access control. This walkthrough uses Amazon Cognito authentication and a serverless AWS Lambda backend to generate scoped embed URLs,…
Seeing what the system did, and what happens after it breaks.
Learn how to build an AI agent once and reuse it across IT, HR, and ops teams, with shared governance, identity, and observability.
OpenReplay replaced Postgres with ClickHouse to ingest billions of events and up to 6 TB a day on a single, vertically scaled instance, with sub-second queries and retention handled by a single TTL setting.
Patch the Planet connects security researchers, maintainers, and incident responders to help address critical vulnerabilities across open source.
Managed ClickStack now flags unusual values on the charts you already use with built-in z-score anomaly detection, and a faster, sharded scheduler lets teams bring tens of thousands of alerts over from tools like Grafana.
When strikes hit Yandex data centers, many assumed Russian bot operations went dark. Cloudflare data reveals that while search crawling stopped, malicious bots simply migrated to other networks.
We redesigned Cloudflare Radar to make real-time global traffic and outage data accessible to a wider audience — including journalists, less technical researchers, and everyday users. By introducing an interactive map and standardized…
We redesigned Cloudflare Radar to make real-time global traffic and outage data accessible to a wider audience — including journalists, less technical researchers, and everyday users. By introducing an interactive map and standardized…
TIL: Using Parseable with Datasette for OpenTelemetry traces I saw Parseable in a Show HN today - it's a new observability platform with both an open source (AGPL) Rust implementation (a single ~180MB binary), an "Enterprise" version…
AbstractIFCO runs one of the world's largest reusable packaging pools with hundreds of millions of crates and pallets...
Vercel Sandbox observability now includes memory usage data. You can access sandbox memory usage data in the dashboard and through the CLI via the vercel metrics command. Sandbox observability memory in the dashboard The Memory Usage…
Release: datasette 1.0a41 Alec Garcia added support for OpenTelemetry to Datasette in this release. I've also refactored all of Datasette's modal dialogs to a single Web Component, which is now documented for other plugins to use .…
What is an AI dark factory? Learn what lights-out manufacturing teaches us about autonomous software development, verification, observability, and AI agents.
ClickHouse, a leader in real-time analytics, data warehousing, observability, and AI/ML, today announced the appointment of Michael Scarpelli to its Board of Directors
Posts from the watchlist that no topic above claims.
Release: battledrones v14: Compass Rose, a four player map I added four player online mode to Battledrones - previous maps were limited to two. Here's the initial prompt I used: Add a new map for a four player battle - four way…
You can now view and manage your team's Spend Management budget from the Vercel CLI. Run vercel budgets and you, or the scripts and agents you run, get back your current spend and the active budget. View, set, and remove a budget This…
Tool: Quick Study Prompt to Claude Opus 5.5 High : Build an artifact where you open a photo and it rapidly (in a few seconds) paints a version of that photo using a canvas element and simulated paintbrush See also my previous experiment…
Introducing Kimi K3.1 Pro Kimi K3.1 Pro landed on the Kimi API today - it's not yet open weights, but Moonshot promise "Full model weights will be released by Nov 4, 2026". They're also offering a "Highspeed" version that's 2-3x faster,…
Cloudflare's new post-quantum GeoKey Manager protects geographically fenced decryption keys from harvest now, decrypt later attacks. Learn how we redesigned our cryptography stack to support strict data sovereignty in a post-quantum world.
Amazon Quick Sight layer maps now support Point and LineString geometry from the same GeoJSON workflow, so you can plot store locations and the routes between them. This post covers the dataset and GeoJSON structure layer maps require,…
AI models are solving more mathematical problems. A pattern is emerging in their successes that could help us predict which problems they’ll solve next.
the Cyber onslaught continues.
Interfaze on Vercel 50 million requests a month Backend running on Fluid compute Multiple model providers behind one fallback layer with AI Gateway Interfaze builds models for deterministic work that general-purpose LLMs handle poorly:…
The Vercel CLI now includes vercel feedback , which sends bugs, friction reports, and feature requests to the Vercel team from your terminal. Coding agents can also send feedback when they run into issues. Use vercel feedback with -f or…
Every Pro team can claim a free domain for one year. That offer now includes .co , .studio , .company , .info , .pro , .live , .digital , and .agency , bringing the eligible TLDs to sixteen. The offer still covers one domain per team,…
DE STAAT - FREEDOM (Official Music Video) OK well this is a first for me... at 00:44 in this new music video by Dutch alternative rock band De Staat : The visual effects are credited to Matties Grooten of Fake it or Leave it . Via flir…
BackgroundAccording to the Communications Fraud Control Association's 2025 survey...
The Amazon Polly Generative engine gets a major upgrade: more expressive, more accurate speech with low latency, plus 14 new voices across 9 locales and general availability of Hindi and Finnish. This drop-in upgrade requires no code…
Amazon Quick is a HIPAA-eligible service, but eligibility is not the same as compliance. This post explains who HIPAA applies to, how PHI flows through Amazon Quick, what the platform provides for regulated environments, and a…
AI developers aren’t just running a single model anymore—they’re orchestrating multiple models in parallel, chaining inference calls, and experimenting...
As fun as AI is, we built so much bad software over the last 30 years, and now the universe is punishing us for our sins. — Armin Ronacher Tags: security , generative-ai , armin-ronacher , ai , ai-security-research , llms
How Moshi turns market data into actionable insights, and why AI tools have to fit the trader’s workflow.
How Moshi turns market data into actionable insights, and why AI tools have to fit the trader’s workflow.
How Moshi turns market data into actionable insights, and why AI tools have to fit the trader’s workflow.
Cloudflare has achieved South Korea's ISMS certification following an audit across 80 controls. Learn how we localized user consent flows, adapted social sign-up paths at the edge, and validated our global security program.
Learn how mortgage document automation transforms loan processing, from document ingestion and extraction to validation and system integration.
Learn how KYC automation replaces manual verification with scalable, compliant workflows that cut costs, reduce errors, and speed up onboarding.
Learn how unstructured data extraction turns documents, PDFs, and text into structured insights using AI, NLP, and LLMs for scalable data processing.
OCR for tables converts complex document layouts into structured, machine-readable data. Learn how LlamaParse preserves table integrity.
OCR for images helps convert photos, labels, and screenshots into structured text. Compare the top AI OCR tools and learn what makes a reliable image-to-text system.
The Big Hardening begins
The Big Hardening begins
With our UK team expanding by more than 200% in six months, we’ve opened a new office at The Gilbert in Finsbury Square and brought in senior leaders to…
What forward deployed engineering means at OpenAI
Learn how OCR for accounts payable automates invoice processing, improves accuracy, reduces costs, and integrates structured data into ERP systems.
Fifteen years ago, TLS was something you'd use to protect the most sensitive areas of your app. Then Google, Let's Encrypt, the EFF, and others led the push for HTTPS everywhere, making encryption the default for every request. Without…
Uber is rolling out ChatGPT Work to more than 30,000 employees, helping teams research, analyze data, create presentations, and complete complex workflows.
You can now deploy apps to Vercel without an account using Vercel CLI . Share prototypes faster with a live URL before signing in to claim your project. From a project directory that isn't linked to Vercel, run: The CLI builds your…
How OpenAI works with distribution partners to uphold consistent safeguards, support responsible access, and clarify expectations in our Usage Policies.
U.S. Hunger, a nonprofit addressing food insecurity, uses Amazon Quick to connect organizational data with leadership decisions. Leaders explore applicant data directly, share KPI definitions and financial context in Quick Spaces, and…
AT&T is working with OpenAI to make legal knowledge easier to access. Learn how its LegalEdge team is rethinking intake and scaling legal judgment.
mammoth-1 and elephant-1 are decision models trained by our internal research team, Base Labs. They are open-source and free to try for a week.
Panache — RTS Belle Époque pour mobile Well this is delightful: I released Battle Drones yesterday under an Apache 2.0 License and there's already a French language fork which reskins the game to take place in steampunk 1900s France and…
From local businesses to global brands, more advertisers can now reach people as they explore, compare and decide what to buy in ChatGPT.
AI factories turn energy into intelligence. But before an AI workload can produce its first token, operators must bring an interconnected system of compute,...
Listen now (44 mins) | Paying for outcomes, building with agents, and the last mile of enterprise AI adoption
Paying for outcomes, building with agents, and the last mile of enterprise AI adoption
Build lightning-fast multimodal AI apps with Gemma 4 on Cerebras. Learn image understanding, vision workflows, and high-speed inference for developers.
Deployments that Vercel automatically skips in a monorepo no longer count against your deployment rate limits. Teams with monorepos now get more deployments from the same limits. Vercel skips a project's deployment when its source code…
Diffusion models have advanced rapidly in recent years, enabling high-quality video generation across a wide range of creative scenarios. As visual quality continues to improve, however, the field is increasingly shifting its focus from…
OpenAI outlines a blueprint for empowering cyber defenders with frontier AI, stronger safeguards, modernized cybersecurity, and more resilient public institutions.
eMudhra Intermediate CA Compromise This is very, very bad news: The private keys of eight eMudhra intermediate CA certificates have been compromised. They chain to roots trusted by the Mozilla, Microsoft, Apple and Chrome root programs…
Discover why AI loops require verification to avoid spiraling. See how Cerebras runs Gemma 4 at 1,500 tokens/sec for fast, autonomous visual loops.
Reasoning boosts AI accuracy, but at a steep cost. Explore test-time compute, agent performance, speed tradeoffs, and when thinking hurts.
Kimi K2.6 on Cerebras matches Gemini 3.5 Flash in intelligence while delivering 5× faster output, lower latency, and open-weight flexibility.
Following the discovery that eight eMudhra intermediate certificates use weak keys, Cloudflare is deploying mitigations and releasing client updates. We explain the impact, our response, and what customers and developers should do next.
WebSocket support in Vercel Functions is now generally available. Since the public beta launched in late June , Vercel CDN has served 300 million WebSocket connections, and now serves 15 million per day. Customers use WebSockets for AI…
My latest vibe coded game experiment is a Command & Conquer style RTS (Real Time Strategy) game called Battle Drones ( play it here ). Unlike my previous experiments , this one is actually fun to play . I'm somewhat embarrassed to admit…
From GPU trading to physical settlement, independent markets, and programmable compute
From GPU trading to physical settlement, independent markets, and programmable compute
Enterprise supply chains are data-rich but insight-poor, with the data to decide scattered across ERP, monitoring, procurement, and email. This post shows how Amazon Bedrock AgentCore exposes those systems as a governed tool layer, and…
Octave built a secure, multi-tenant AI foundation on Amazon Bedrock, using Cohere Embed v4 for embeddings and Amazon Bedrock AgentCore for isolated agent execution, to power a Complaint Triage Assistant that searches 2 billion regulated…
Western open models strike back
Whether you are iterating on a quick bug fix or planning a complex migration, different tasks call for different models. In keeping with Google Cloud's commitment to an open ecosystem and to give developers the right model for every…
Cloudflare is collaborating on tsrs, a Rust port of TypeScript that is up to 8x faster and uses 70% less memory.
An independent Schlotzsky’s and Cinnabon franchisee cut menu repricing time by 95% and built a workflow that could free up $100K in cash flow.
Terminal X’s Eval Agent grades outputs against a 54-criterion rubric, catching 25% more deviances from a fund’s mandate with GPT-6.1 Sol.
Congrats to DeepSeek V4.1 Flash 1017, MiniMax M3.1 Flash, and Mercury 3 (details below) but they’re not going to be in super wide use just yet.
Congrats to DeepSeek V4.1 Flash 1017, MiniMax M3.1 Flash, and Mercury 3 (details below) but they’re not going to be in super wide use just yet.
OpenAI is acquiring Orchestra Bio to bring AI-powered operations to life sciences R&D, helping biopharma teams accelerate drug development and advance new treatments.
Fireworks Free Weights Club kicks off with Muse Glimmer: better predictions on Kalshi, a reproducible training recipe, and free SFT through November 30.
New Amazon SageMaker AI customers can now land in a fully configured SageMaker Studio environment in only a few steps. Sign up with an identity you already have, such as Amazon, Apple, GitHub, or Google, with no advanced AWS account or…
While almost every platform now supports open formats, what makes a catalog truly...
Amazon SageMaker JumpStart onboarded over 100 new models from 30 providers in the first three quarters of 2026. This recap covers the model-onboarding highlights and JumpStart enhancements, including the simplified Model Hub,…
We are updating the Terms of Service for Hobby and Pro customers, effective October 19, 2026. The updates clarify some existing Vercel practices, expand terms on topics such as managed accounts and credits, and reflect applicable legal…
Clairic is a new model from Cloudflare that is a System One extraction modeI, capable of extracting, computing, estimating, and forecasting data from unstructured inputs.
TwelveLabs Pegasus 1.5 is now available in Amazon Bedrock. The video language model turns long-form video into structured, timestamped metadata: define the moments that matter and get every matching segment back as JSON, with new video…
Deepgram worked with the AWS Forward Deployed Engineering team to port its Nova-3 speech-to-text model to AWS Trainium2 for offline batch inference. This post covers porting to the AWS Neuron stack, kernel-level tuning, a data-parallel…
Generative Recommenders (GenRecs) represent a paradigm shift from scoring every ad to reasoning about ad content and user preferences together in order to predict the best outcome for advertisers. Leveraging the world knowledge built…
How Numan built Nu, an AI health coach, on LangGraph and LangSmith with per-category evals, simulated conversations, and safety gates on a regulatory line.
With the recent release of the Muse family of foundational models, we want to take you inside the infrastructure that has made it all possible. The first phase of Prometheus, our gigawatt-scale supercluster, is already complete and…
Oxide and L Labs set 18 STAC-M3 world records on a single Oxide virtual machine, opening a partnership to bring L to firms that run tick data analytics on-premises.
CVE-2026-88771 lets attackers run commands on NetScaler Gateway before anyone signs in. Patching takes time most teams don't have, so the sign-in decision has to move in front of the vulnerable code.
Query Insights for ClickHouse Managed Postgres now shows what active sessions are waiting on, with wait event breakdowns, blocked-session and lock-wait charts, and the same metrics in Prometheus.
Build agents that automate tasks across public and private websites with managed browsers, native toolbox integration, Live View, and human control. The post Browser automation tool in Microsoft Foundry is now GA appeared first on…
You can now deploy static sites with Vercel Drop without an account. Drag and drop a file, folder, or ZIP archive to get a live URL. Deployments without an account expire after one hour. To keep your site live, sign in or create an…
484 older posts in this topic are not shown; the page lists the 100 most recent of each topic. Narrow the window to bring them within the cap.
Grouping by subject is the point. A reader gives you 45 separate streams and leaves you to notice that four companies wrote about the same scheduling problem this week; this page puts those four posts under one heading. Topics come from the filter terms already listed on the blogs-to-watch page, so the vocabulary is the course's.
Each post is scored by keyword against every topic, using its title and its summary, and it is filed under the topic it scores highest against. A term that names a subject on its own counts for more than one that merely co-occurs with it, and a match in the title counts for three times a match in the summary. Any second topic a post also matches is shown as a label beside it. The method is keyword matching rather than a model, which makes it predictable, cheap, and occasionally wrong.
Nothing on this page is required reading.
The index covers 45 sources from the watchlist. 37 publish a feed and are read from it; the other 8 publish none, so their index pages are scraped and each new post's own page supplies the title and date its card omits.
Last run finished 26 Oct 2026 at 01:07 UTC. The index holds 1500 posts, keeps them for 120 days, and shows 1500 posts on this page.
1 source failed on the last run. Replit (HTTP 403)
1 source answered but yielded no post: Anthropic Engineering.
The last run read 824 posts across every source and added nothing new.
4 sources on the watchlist are not aggregated here, so check them by hand.
A date shown as "first seen" is not a publication date. Some sources publish no date at all, so the page records when the post entered the index instead of guessing.