9 results for “GPU”
SenseTime launches the Galaxy Project with nearly 20 partners to scale domestic AI chip infrastructure in China, claiming 25x token throughput growth by Q4 2026 and a heterogeneous hybrid inference stack across seven domestic chip vendors.
Read more →
SpaceX has placed a $52 billion order with Foxconn for nearly 1 million Nvidia GB300 GPUs, marking one of the largest AI infrastructure investments. Deliveries are expected from Q4 2026 to Q1 2027.
Read more →
AMD launches Helios AI rack-scale system with 72 MI455X GPUs and 18 EPYC Venice CPUs, claiming 30% better tokens-per-dollar than Nvidia. OpenAI, Anthropic, Meta, and Microsoft commit to gigawatt-scale deployments starting Q4 2026.
Read more →
Nvidia's open-source Medical Physics Simulation framework combines classical physics and generative AI to train surgical robots at scale, addressing the data scarcity that has slowed healthcare robotics deployment.
Read more →
AMD shares surged after Microsoft committed to deploying AMD's Helios Rackscale solution at scale on Azure, validating AMD's AI hardware ahead of its Advancing AI 2026 conference.
Read more →
Moonshot AI launched Kimi K3, a 2.8-trillion-parameter open 3T-class model with a 1-million-token context window and native multimodal training. The model is live on kimi.com and via API; full weights are scheduled for July 27, 2026.
Read more →
Nokia launched the industry's first commercial AI-RAN platform on July 15, built on its anyRAN software and NVIDIA's Aerial system, delivering over 20% spectral efficiency gains with a roadmap to double capacity by 2028 through a software subscription model.
Read more →
Meta is reportedly in talks to lease AI computing capacity to Anthropic in a deal worth up to $10 billion over two years, according to The New York Times, marking the latest cross-competitor compute arrangement driven by the AI hardware shortage.
Read more →
NVIDIA has introduced the Jetson Thor T3000 and T2000 modules, giving robotics developers new options for running multimodal AI models and agent workloads directly at the edge.
Read more →