Articles

# AMD Lemonade：支持 GPU 和 NPU 的开源本地 AI 服务器 AMD 发布了 **Lemonade**，一款快速、开源的本地 AI 服务器，可在 GPU 和 NPU 上运行文本、图像和语音模型。该工具代表了 AMD 向本地 AI 推理市场推进的举措，挑战 NVIDIA CUDA 生态系统的主导地位。 ## 什么是 Lemonade？ Lemonade 是一个轻量级 AI

# AMD Lemonade: Open Source Local AI Server Supporting GPU and NPU AMD has released **Lemonade**, a fast, open-source local AI server that runs text, image, and speech models on both GPUs and NPUs. T

Hypura is a new open source storage-tier-aware LLM inference scheduler for Apple Silicon that optimizes model data movement between RAM and storage, enabling larger models to run efficiently on memory

George Hotz and the tinygrad team have begun shipping the Tinybox — a personal AI inference computer capable of running 120B+ parameter models entirely offline. Red V2 starts at $12,000, Green V2 Blac