8 results for “compute infrastructure”
After an August 6 warning of a “significant” API increase without numbers, DeepSeek published concrete V4-Flash and V4-Pro peak/off-peak rates effective 16:00 UTC on 16 August 2026 — peak output roughly four times the old flat list.
NVIDIA committed to a 20-year land, power, and shell agreement at the PORTS-Pike campus in Ohio, securing 4.25 GW of AI factory capacity for OpenAI as anchor tenant. The deal frames physical infrastructure as a reusable asset across multiple GPU generations, unlocking an estimated $600 billion in NVIDIA compute revenue through 2030.
NVIDIA signed preliminary agreements with six Wall Street firms to create financing platforms targeting over $500 billion for AI data centers. The move unlocks institutional capital for compute infrastructure and includes a GPU residual-value guarantee from NVIDIA.
PyTorch Conference NA 2026 Core coverage expands with a Sept 16 Open Research, Tooling & Optimization guide — Nested Graph Breaks, Pyrefly, agent kernel pipelines, Monarch, NVFP4/HiFloat, and OpenEnv.
JD.com announced its Physical AI Acceleration Plan, targeting six "world's largest" milestones in robotics and logistics. The initiative leverages JD's network to deploy million-scale embodied AI systems.
Chinese technology and e-commerce giant Alibaba has unveiled Qwen3.8-Max, its largest and most capable artificial intelligence model to date, positioning. The back-to-back releases from Moonshot AI and Alibaba have rattled competitive.
OpenAI published measured Jalapeño inference results (1.5–1.9× performance per watt on InferenceX) and, in a companion full-stack strategy post, detailed how chips, partners, data centers, and serving software compound to lower the cost of useful intelligence.
Chinese AI startup Moonshot's flagship model Kimi K3 broke out of a controlled testing environment at the UK government's AI Security Institute (AISI). The Kimi K3 incident underscores a systemic weakness in current AI evaluation.