AI Infrastructure news and analysis on Web Pulse: 39078 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
Broadcom Inc. is expanding its VMware portfolio with new capabilities for deploying AI workloads, including AI Infrastructure Quickstart, which consolidates vCenter cluster deployment, vSphere Supervisor activation and V…
Vercel's AI Gateway now serves Moonshot AI's Kimi K3.1 Pro and Kimi K3.1 Pro Highspeed through additional Zero Data Retention (ZDR) providers, where Moonshot AI had been the only provider for both models. Both models off…
A two-person lab released Linum v3 text-to-image model weights under Apache 2.0 after training the model for roughly $20K on 660M samples using its Pyramid-JiT pixel-space diffusion architecture. The lab says dropping th…
OpenAI built Pando, a Rust-based Git-compatible backend that replaced the storage layer behind its monorepo without changing the protocol or client, after reliability fell below 90% and commits landing on master grew 2,4…
A reference implementation built on the Strands Agents SDK and Amazon Bedrock AgentCore cut the tokens each script adds to context by about 95 percent through a single write-and-execute tool, and shrank the system prompt…
A developer built a no-sign-up image-to-video tool that turns a single photo into a 5-second AI clip, running on Cloudflare Workers with D1 for state, R2 for storage, a Cloudflare Container running FFmpeg, and cron trigg…
GPUWorks, released by flitai-io as a 40 MB control-plane download, is a GPU cluster manager and Model-as-a-Service platform that deploys vLLM, SGLang or template-defined inference engines across NVIDIA, AMD Instinct and …
Nvidia's RTX Spark laptop runs AI inference at 13.9 tokens per second on native Linux wall power versus 65.4 tokens per second on Windows, a 79% drop documented in a MindStudio head-to-head benchmark on GPT-OSS-120B repo…
Google Compute Engine's API multi-versioning is now generally available, adding immutable date-based API snapshots alongside the existing /v1 endpoint, with the first stable version, 2026-09-01, due on a quarterly cadenc…
At the SEC 2026 Minimalist Edge AI Challenge hosted at Santa Clara University, 17 teams from 7 universities deployed computer vision models on compact boards including the Arduino UNO Q and Qualcomm development hardware.…
Confluent released its Q4'26 Data Streaming Platform update, adding Query as a fifth DSP pillar alongside Stream, Connect, Process, and Govern, with Lightning Tables at its center to make Kafka streams queryable in milli…
ZStack open-sourced GPUWorks, a lightweight enterprise AI inference and Model-as-a-Service platform, on November 4, 2026 under the Apache License 2.0 in the flitai-io/gpuworks repository. In a reported test, one GPUWorks…
A developer deployed a private model gateway on AWS over two weeks to test whether a centralized layer between agents and models can govern access, spending, data policy, reliability and visibility across ten agents and …
A TIL post outlines six techniques for reducing large language model inference costs: prompt caching, model routing, batch APIs, semantic caching, context trimming and smaller or distilled models. The post cites Anthropi…
A developer at wrightfunctions.com outlined four techniques for cutting LLM inference costs by routing requests between cheap and expensive models rather than sending every query to the priciest one. The writeup cites An…
Restate added durable coroutines to its TypeScript SDK, giving persistent, concurrent, multi-step routines the same persistence and recovery semantics as workflows at only milliseconds of overhead per durable action rath…
STMicroelectronics is expanding its STM32 microcontroller roadmap with multicore devices, hardware accelerators and AI-assisted development tools, targeting the cloud AI and physical AI markets, Remi El-Ouazzane, preside…
Anthropic's Frontier Red Team open-sourced FlashProve on November 2, 2026, a research prototype that generates zkSNARK proofs binding an output token sequence to a specific committed set of model weights without revealin…
Kong introduced AI Management in Kong Konnect, a set of capabilities in one control plane that addresses four problems platform teams face running agents in production: unattributable spend, unmanageable credentials, API…
Kong released Kong AI Observability, a product that traces a full AI task across models, tools, agents and policies within the connectivity path rather than assembling telemetry from separate systems, according to a post…