Nvidia invested $2B in Marvell for NVLink Fusion partnership

NVIDIA paying half the capital expenditure to expand supplier fabs for critical thin-film material.

Announced Nemotron 3 Nano Omni, an open multimodal model processing video, audio, images, and text.

Intel demonstrated UCIe-S interconnect at ISSCC, beating 3nm EMIB with 3x higher data rate

Nvidia invests $2 billion in Marvell Technology for NVLink Fusion interconnect development

Nvidia trained a billion-parameter LLM without backpropagation or full-precision weights, using zero gradients.

Open-sourced Kimono, a motion diffusion model for humanoid robots

Intel and SambaNova Systems jointly proposed a hybrid inference architecture blueprint for agentic AI workloads.

Announced it is joining the 'Terafab' project consortium with SpaceX, xAI, and Tesla to refactor silicon fabrication technology.