Nebius Group NV, a Dutch operator of artificial intelligence data centers, today announced plans to buy software maker Eigen ...

4月2日，国内量子计算团队正式宣布开源QLLVM——国内首个打通经典高性能和量子混合编译链路的编译框架。该项目实现了对多种量子编程语言和量子计算后端的统一支持，能将量子程序与CUDA、C++程序混合编译，为量子计算提供了工业级的编译基础设施。

“时代洪水冲天下，长空激浪正当时。”在日前举办的昇腾生态媒体沟通会上，华为昇腾专家以此作结，系统性地向外界披露了昇腾在生态建设上的最新进展与核心战略。 CANN的底层重构 作为昇腾AI异构计算架构，CANN是连接硬件与上层应用的核心纽带，其性能与易用性直接决定昇腾生态的竞争力。

DESILO, a pioneering deep-tech company specializing in privacy-enhancing technologies, has announced the release of the world's first Fully Homomorphic Encryption (FHE) library to seamlessly integrate ...

这是英伟达CEO黄仁勋近期在一档播客节目中发出的警告。让他发出警告的对象，是即将发布新模型的中国AI公司DeepSeek。

DeepSeek V4 模型以 CUDA + Tilelang 组合方式使用了数十个算子。为了能在多种 AI 芯片上运行 DeepSeek V4，众智 FlagOS 采用 Triton/Triton-TLE 重写了全部新增算子，并基于 ...

Flexible, power-efficient AI acceleration enables enterprises to deploy advanced workloads without disrupting existing data ...

Graphics processing units have fundamentally reshaped how professionals across numerous disciplines approach demanding ...

Build AI hackathon projects on AMD MI300X GPUs with $100 in free credits, ROCm open-source stack, and free courses from the ...

如果你关注科技新闻，一定经常看到“Python+AI”这样的组合。从ChatGPT的横空出世，到Midjourney生成的惊艳图像，再到自动驾驶汽车的感知系统，这些前沿技术的背后，几乎都有Python的身影。

NVIDIA's new cuda.compute library topped GPU MODE benchmarks, delivering CUDA C++ performance through pure Python with 2-4x speedups over custom kernels. NVIDIA's CCCL team just demonstrated that ...