2026 年 10 月 8 日，Google AI Edge 团队正式开源 ML Drift。这是一套面向端侧 AI/机器学习推理的 GPU 计算引擎，采用 Apache 2.0 许可证；它既是 LiteRT 新一代 GPU 加速能力的重要组成部分，也可以通过独立 C++ 库集成到其他推理系统中。

Google's open source ML Drift engine accelerates AI on everything from smartphones to desktops while simplifying support for ...

A new open-source solver called GAUZZ runs geodynamic Stokes equations on GPUs up to eleven times faster than a 32-core CPU ...

Executive SummaryIn this week's AI product and feature news, the AI competition has expanded further from "model performance" ...

PhotoPrism's October 2026 update improves AI image recognition, adds NVIDIA GPU acceleration, and enhances face detection and ...

An independent lab benchmark published October 1, 2026, showed that running F5's AI-aware load balancer natively on NVIDIA BlueField-3 Data Processing Units (DPUs) — rather than on host CPUs alongside ...

AMD has detailed a new ray tracing technique called tetrahedral cages, and according to a blog post on the company's GPUOpen developer site, it can cut BVH memory use by roughly 47x while speeding up ...

IT之家9 月 22 日消息，龙芯中科今日宣布推出支持自研通用 GPU 的龙芯加速计算平台（Loongson Accelerated Computing Platform）首个软件版本，面向龙芯通用 GPU 提供从底层驱动、编程环境到 AI 模型推理的 ...

Overview:  Deep learning uses multi-layer neural networks to learn patterns from data.CNNs, RNNs, LSTMs, transformers, and ...

If you were around during what many consider the golden age of PC gaming, back when SLI was still possible, and, going back even further, when gamers put dedicated PhysX cards into their PCs, the ...

As part of its Advancing AI launch event last week, AMD unveiled ROCm.ai, a new “developer experience” that introduces AI coding agents into AMD’s ROCm development tools for GPUs. According to AMD, ...

Yangqing Jia, the AI researcher who created the Caffe deep learning framework and later co-founded LeptonAI, has departed NVIDIA — roughly 14 months after the chipmaker acquired his startup for ...