Tag
AI cloud provider Lambda is raising up to $4 billion at a $14.5 billion pre-money valuation in what may be its final private round before a planned 2027 IPO, with Coatue and Blackstone leading. Much of Lambda's ballooning backlog appears driven by a $35 billion commitment from Anthropic, raising questions about concentration risk.
China's AI giants are undergoing a dramatic talent shift: post-training algorithm leads brought in from Silicon Valley are being sidelined, while AI Infra engineers who can orchestrate ten-thousand-GPU clusters are being courted with multi-million-yuan salaries, marking the field's pivot from algorithm heroics to infrastructure-scale engineering.
FT报道腾讯与Oracle签署5年期AI芯片租约，约10万颗先进AI芯片、价值约70亿美元，部署在东南亚的Oracle数据中心且无法进入中国，凸显美国出口管制只约束物理芯片流向而非算力租赁。
Lila Sciences presented at Ray Summit about their internal self-service AI research platform, which uses Flight, Ray, and GitOps to support automated scientific discovery through optimized build-test-learn loops.
The article explores using a ZIMA Board 2 with an RTX 2000 ADA GPU as an affordable self-contained setup for running the Qwen 3.8 27b AI model, comparing it with alternatives like the Mac Mini M5.
AMD releases ROCm 10.0, a major update to its open-source GPU compute platform, marking a decade and introducing native agentic AI developer experience with ROCm.AI.
This paper analyzes NLP conference papers from 2020 to 2025 to examine the relationship between reported GPU resources and scholarly impact, finding that while resources are associated with higher citations, they explain little of the variance in impact.
Nvidia is partnering with major financial firms to assemble $500 billion in financing to establish GPU compute as an investable asset class, drawing comparisons to mortgage-backed securities but facing skepticism about oversaturation and profitability in the AI industry.
NVIDIA announced partnerships with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR to establish financing platforms that could mobilize over $500 billion for AI infrastructure, positioning AI factories as a new investable asset class.
A MIT Technology Review piece explores how AI professors are adapting to a research landscape dominated by private frontier labs, grappling with GPU costs, restricted access to model internals, and shrinking federal funding while focusing on socially relevant questions that profit-driven companies ignore.
slurpjson is a Rust library that parses JSON entirely on the GPU using wgpu compute shaders, decomposing parsing into parallel prefix scans for research purposes.
A free interactive book teaching graphics programming with WebGPU in JavaScript, covering from basics to advanced topics like GPU compute and Gaussian splatting.
Computable launches a marketplace to buy, sell, and redeem GPU hours by the week with instant liquidity, founded by ex-Jump Trading and Coinbase engineers and backed by Y Combinator.
AMD has tagged the ROCm 7.14 'TheRock' tech preview, bringing AI training enhancements, performance improvements up to 16% for select AI workloads like Comfy UI, and ongoing Windows support for the open-source GPU compute stack.
Discusses methods to run CUDA software on non-Nvidia GPUs, enabling use of Nvidia's AI ecosystem on other hardware.
Hugging Face Storage is now a first-class backend for SkyPilot, allowing users to mount Hugging Face repos and buckets into jobs on any cloud with zero egress fees, enabling flexible GPU compute across providers.
ZLUDA 6 is released, enabling unmodified CUDA applications to run on non-NVIDIA GPUs with new support for PhysX and Blender.
This paper introduces Computable Fair Division (CFD), a framework using Boltzmann-Softmax control to balance efficiency and fairness in AI resource allocation, with real-time adaptation via AHC++.
The article estimates that daily token demand in China has reached 1,000 trillion, but computing capacity remains inadequate, leading to a supply-demand imbalance.
The Swiss AI Initiative, launched in December 2023 with over 10m GPU hours and 20m CHF funding, is a major open science effort for developing AI foundation models involving 800+ researchers across Swiss institutions. Backed by the Alps supercomputer and collaborative support from ETH and EPFL, it aims to provide transparent models and datasets for Swiss stakeholders.