AI Infrastructure news and analysis on Web Pulse: 35328 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
Slotstream, a new Swift binary, lets users run the 104 GB Qwen3.8-Flash-Next model on a 48 GB Mac at about 12 tokens per second by streaming weights from SSD, with peak memory of 32 GB and a cold start to first token in …
Anthropic's Claude Fable 5.1, its most capable model to date, is now available on Snowflake Cortex AI in private preview, offering same-day availability as a launch partner. The model delivers capability gains in agentic…
Patent filings at the intersection of AI and semiconductors increased 114% over the past five years, outpacing the broader sector's 78% growth, according to a report by IP management provider Anaqua. The report highlight…
Empirik, an AI-powered infrastructure observability startup incubated by Sequoia Capital, launched Tuesday with $21 million in seed funding from Sequoia, Canapi, and Alumni Ventures. The company, led by CEO Kartik Chandr…
Prashant Jalan, a former Google applied AI lead, launched Guickly on September 1st with $4.2M in seed funding led by Engineering Capital to help enterprises track and control AI spending. The company's software shows whi…
Amendable launched a managed Git repository service designed for AI agents, allowing developers to provision private repos via API with scoped access tokens and standard HTTPS Git. The service includes webhooks, a web UI…
Orchestra, an AI-native platform for enterprise data and AI workflows, formally launched with $4.6 million in total funding, including a $3.3 million seed round led by Differential Ventures and a $1.3 million pre-seed ro…
A three-week operational field report from Python Visuals, running on the Apeiron OS substrate, found that its 'independent' AI reviewers were all the same underlying model, undermining the validity of cross-vendor evalu…
President Donald Trump defended AI data center development on Truth Social, warning that communities opposing the projects risk becoming 'backwards and poor' and ceding ground to China in the AI race. A Gallup survey in …
Atos partnered with AWS to run an agentic AI League event that upskilled 400 engineers over three days, moving them from limited hands-on experience to building multi-agent systems with pathfinding, guardrails, memory, a…
Shaide, a Kubernetes-native AI platform for distributed multi-model inference, has been released under the Apache 2.0 license. It uses vLLM for inference, llm-d for orchestration, KV-cache-aware scheduling, and an intern…
Anthropic shipped three new API endpoints on August 11 that let enterprise security teams pull full transcripts of every Claude Code and Claude Cowork session running on developer machines, capturing prompts, bash comman…
The Coalition for Secure AI (CoSAI), an OASIS Open Project, hosted a hybrid working session on August 28 to advance the Open Delegation & Identity Standard (ODIS), a vendor-neutral architecture for cryptographic identity…
The Pragmatic Engineer has launched an essay competition with a grand prize of $10,000 and additional prizes of $2,500, inviting software engineers to submit essays of 3,000 to 10,000 words on how AI and LLMs are changin…
MinIO is partnering with Saudi Arabia's sovereign wealth fund-owned HUMAIN to build the data foundation for a sovereign Saudi AI Stack, with MinIO named HUMAIN's data and memory foundation partner. The collaboration incl…
Jamf, which manages Apple devices for over 76,000 organizations, built a real-time spend enforcement system for Amazon Bedrock that tracks each engineer's daily AI costs and automatically restricts access to premium mode…
DE-CIX expanded Microsoft Azure ExpressRoute Metro connectivity across New York, Frankfurt, Madrid, Amsterdam, and Singapore, adding dual-homed private connections to support high-availability AI workloads. The operator …
A 390-line Python script benchmarked six AI models via DigitalOcean's OpenAI-compatible inference endpoint and found that the cheapest model, Mistral 3 14B, delivered the fastest time-to-first-token at 1250 ms, while the…
Cisco ThousandEyes built OdysseyCD, an internal continuous delivery platform on Kargo, to replace manual release cycles with a centralized, policy-driven promotion model, now managing 137,000+ release gate evaluations ev…
Fal, the San Francisco-based generative media platform co-founded by Burkay Gur and Gorkem Yurtseven, released H3 Max on September 1, a post-trained video model based on MiniMax's open-weights H3 that generates five-seco…