You’re offline. Open downloads
CUDA
Automatically indexed from the existing article tag.
Theses, forecasts, models, and paths
No published research objects yet
This topic is ready for editors to connect a thesis, forecast, model, or research path.
Latest articles
Article · related
Nvidia Turns Rubin HPC Into a Scientific-AI Compute Roadmap Test
Nvidia announced Rubin HPC at SC26 on November 16, 2026, positioning the Vera Rubin-family GPU for simulation, data processing, science, engineering and analytics workloads. The company disclosed more than 50,000 CUDA cores, sparse-data acceleration, seventh-generation NVLink and expected 2028 availability. Independent technical coverage framed the move as a renewed push into HPC while noting that memory, power, pricing and full precision-performance details remain open.
Published Nov 17, 2026
Article · related
DeepSeek Turns Huawei Ascend Tools Into a CUDA-Bypass Stack
DeepSeek said on September 30, 2026 that it partnered with Huawei to build programming infrastructure for Huawei Ascend chips, including open-source compute and communication libraries. The release includes DeepGEMM-Ascend, DeepEP-Ascend, and Ascend support around TileLang, with DeepGEMM-Ascend listing initial Ascend 950 support. The strategic signal is clear: China’s AI hardware bottleneck is becoming a software ecosystem contest against Nvidia’s CUDA-centered moat.
Published Oct 2, 2026Join Pattern Nexus for premium research, the archive, and ad-free reading.