AI Infrastructure news and analysis on Web Pulse: 34317 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
Microsoft Foundry Agent Service has adopted the Model Context Protocol (MCP) as a first-class remote tool type, introducing a Toolbox construct that acts as a governance layer for agent tool calling. The design separates…
A cost-analysis writeup argues that cache economics, not sticker price, determines real AI spending for small businesses, since repeated context tokens can be cached and billed at a fraction of full input rates. Using a …
Huawei Technologies rotating chairman Eric Xu Zhijun said at the Huawei Connect 2026 conference in Shanghai on Thursday that a major domestic shift toward Huawei's Ascend-based AI computing infrastructure for model train…
DeepSeek v2 reduces memory footprint by 437x compared with DeepSeek v1, according to a video report. The source provides no further details on the model, its release date, or the benchmark behind the figure.
A developer analyzed the architecture of open-higgsfield, an open-source multi-model video and image generation studio, detailing how it bypasses Next.js Server Action client-side serialization through a single-flight fa…
A developer outlined production techniques for cutting LLM API costs, arguing that prompt-prefix caching, task-based model routing, tighter context management, batch APIs, and hard usage limits deliver far larger savings…
A scan found 230 of 243 unauthenticated LocalAI instances vulnerable to exploitation, a flaw recently exploited in a large-scale campaign that compromised high-value targets including infrastructure linked to the Thai mi…
A new arXiv paper (arXiv:2609.19213v1) introduces layer-wise curriculum learning for efficient LLM compression, a method that partitions a model into multi-layer segments to transfer knowledge from teacher to student mod…
Researchers introduced block parallelism (BP) and context-sharded block parallelism (CSBP), new distributed parallelism dimensions for training block diffusion language models (BDLMs) on long contexts, reporting throughp…
Researchers introduced radio-frequency convolutional neural networks (RF-CNNs), which repurpose the frequency mixer in existing wireless radios to perform CNN inference, according to a paper published as arXiv:2609.19279…
A new arXiv paper (2609.19499v1) reports that generation schedule, not candidate count alone, determines the energy and latency cost of LLM test-time scaling: on A100 GPUs, eight serial generation calls with one candidat…
A position paper posted to arXiv (2609.19203v1) argues that compound agentic AI systems need a Foundation Model Operating System (FMOS), a system layer that virtualizes foundation model interactions the way virtual machi…
A new arXiv paper (2609.20068v1) unifies marginal utility, matrix factorization, and the transformer Key-Value cache under a single allocation rule that retains the top dimensions whose eigenvalue exceeds the shadow pric…
Nunchux AI researchers Muyang Li and collaborators published the VC-Attention paper on September 14, reporting attention-kernel speedups of 1.46x to 1.59x on NVIDIA H200 and B200 data-center GPUs and up to 1.70x faster e…
VIDRAFT AI Research published a paper (arXiv:2609.20269) testing whether Transformer attention mechanisms need to repeat identically at every layer, using a 7×7 Latin square to spread seven sequence mixers evenly across …
At the AI Infra Summit in Santa Clara, technologists from Amazon Web Services, Oracle, Broadcom, Qualcomm and d-Matrix outlined efforts to optimize AI infrastructure as token costs spiral, with Qualcomm executive vice pr…
Microsoft Chairman and CEO Satya Nadella said India could build the world's largest developer community by 2030, speaking at the IDEAS4INDIA: Indian American Tech CEOs Forum in Seattle hosted by the Consulate General of …
Thailand's Energy Policy and Planning Office forecasts electricity consumption will rise 5.3 percent in 2026, the fastest growth in nine years, driven by electronics manufacturing and data centre investment tied to the g…
A developer quantized a fine-tuned DistilBERT Banking77 support-ticket classifier to ONNX int8, shrinking the model from 256 MB to 64 MB so it could run entirely client-side in the browser via Transformers.js with no bac…
A forum thread is crowdsourcing real-world local LLM deployment data across five VRAM tiers — ≤16GB, 24–32GB, 48–64GB, 96–128GB, and 196–256GB+ — asking users to report model, quantization (AutoRound W4A16, GGUF Q4_K_M/I…