AI Infrastructure news and analysis on Web Pulse: 35675 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
A developer has created a privacy-first personal finance assistant that runs entirely offline in the browser using LiteRT's LM Web API and the Gemma 4 E2B small language model. The application caches model weights in the…
Maritime, a serverless infrastructure platform, announced it can run stateful AI agents in isolated microVMs, enabling one agent per customer with no shared runtimes. The company offers flat pricing starting at $0 for fr…
Applied Intuition, Inc., a Silicon Valley company valued at $15 billion, is hiring a Software Engineer for multi-agent UAV autonomy in its Seoul office to build tools for defense customers developing coordinated unmanned…
Chinese AI lab Z.ai revealed on Wednesday that it is behind the viral anonymous Ox Alpha model, releasing GLM-5.3-Flash, a natively multimodal model with 320 billion total parameters and 18 million active, which outperfo…
Apple's M5 chip, announced October 2025, introduces a Neural Accelerator in every GPU core, making prompt processing 3.3 to 4.1 times faster than the M4, according to Apple's MLX benchmarks. Token generation improves onl…
MyZubster, an Italian open-source digital ecosystem under development, is exploring an architecture that connects real-world data, provenance, AI, and verification to address the challenge of verifying what actually happ…
A developer at forge.workstyle.tech explores cost reduction in GPU cloud services and the rise of containerized data centers, highlighting their advantages in rapid deployment and reduced power consumption. The article n…
NVIDIA Corp. reported $96.2 billion in revenue for the quarter ended July 26, 2026, and guided to $108 billion for the current period, with CEO Jensen Huang stating 'Compute is revenue, and demand is accelerating.' The c…
AWS has launched AgentCore Evaluations, a framework-agnostic evaluation layer for AI agents that uses OpenTelemetry telemetry as the contract. The service scores agents built with LangGraph, LlamaIndex, OpenAI Agents SDK…
A developer argues that AI agents' failures in production are due to memory and state management issues, not reasoning limitations. The post outlines a multi-layered memory architecture—short-term, long-term, and procedu…
AI agents can silently return wrong answers when they read from stale data copies, and the staleness budget should be set per question rather than per system, according to a guide that outlines a framework for classifyin…
Nowledge Labs' Nowledge Mem provides a unified memory hub that lets AI agents and LLM chat tools across multiple devices share persistent context, with the hub storing memories, conversations, and a knowledge graph on a …
AgentBadge's Agent Readiness scanner assigns scores like 87/100 through a deterministic pipeline of rules, evidence, assertions, and scoring, rejecting LLM-based opinions in favor of reproducible measurements. The scanne…
1Password's IT team found that AI adoption outpaced visibility, leading to fragmented spend data across vendor dashboards and inconsistent governance. The company adopted its AI Spend and Consumption Management tool to c…
AMD released ROCm 10, making ROCm.AI generally available, an AI-native developer platform that combines ROCm Hyperloom, AMD Skills, and the ROCm CLI to accelerate AI workload development and optimization on AMD hardware.…
AMD released ROCm 10.0, the first major version bump since the 7.x series, built entirely on TheRock automated build system and introducing ROCm.AI, a native agentic AI developer experience. The release consolidates GPU …
Telnyx has made GLM-5.3-Flash, the first native multimodal model in the GLM-5 series, available on its Inference API, offering stronger intelligence than GLM-5.2 at roughly 92% lower input token cost and 94% lower output…
Solo.io has published open source distributions of kagent and agentgateway on AWS Marketplace, Microsoft Azure Marketplace, and Google Cloud Marketplace, both free to deploy and installable via each cloud's native flow. …
PageIndex SDK now supports fully local reasoning-based RAG workflows, allowing users to build tree indexes, store documents, run retrieval, and chat with long documents on their own machines using their preferred LLM and…
Adobe is hiring a Senior Staff Machine Learning Engineer for its Firefly Foundry media intelligence platform, offering a salary of $239k–346k/yr for an on-site role in San Jose. The engineer will lead data processing, in…