LG Electronics has become the first Korean company to receive Nvidia's final quality certification for a 600kW coolant distribution unit (CDU), a critical component for liquid cooling in AI data cente…
The Apache TVM community introduced TIRx, an open-source, hardware-native DSL and compiler for ML kernels that compiles to GPUs and specialized AI accelerators. TIRx targets fast-moving frontier kerne…
Startup Majestic Labs claims its Prometheus server, built with Ignite AI Processing Units and LPDDR6 memory, can deliver up to 128 TB of unified memory and 25.6 TB/s bandwidth, more than 50 times the …
Nvidia partnered with Mizuho Financial Group on July 15, 2026, to build AI factories using Nvidia's confidential computing technology for secure private AI environments in banking, focusing on fraud d…
Inception Labs' Mercury 2 diffusion language model outperforms Google DeepMind's DiffusionGemma on reasoning benchmarks while achieving parallel text generation speeds of over 1,000 tokens per second.…
MiniMax released its M3 multimodal model on NVIDIA's accelerated infrastructure, offering a free public endpoint via NVIDIA's API catalog. The 428-billion-parameter model processes text, images, and v…
French startup Mistral AI and Nvidia announced a partnership to launch Mistral Compute, a European AI infrastructure stack that will include roughly 18,000 Blackwell GPUs and a planned 40-megawatt dat…
Researchers have developed ThriftAttention, a mixed-precision attention algorithm that selectively computes only 5% of query-key blocks in FP16 precision while processing the remaining 95% in FP4, rec…