{
  "id": 5208793,
  "title": "Lemonade 11.9 Local AI Server Released With Super Exciting AMD ROCm HRX Backend",
  "url": "https://urgent.news/2026/09/03/lemonade-11-9-local-ai-server-released-with-super-exciting-amd-rocm",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-03T00:32:40.000Z",
  "source": {
    "name": "Phoronix",
    "slug": "phoronix",
    "url": "https://www.phoronix.com/news/Lemonade-11.9-ROCm-HRX"
  },
  "original_language": "en",
  "account": "AMD has introduced a new AI server, Lemonade 11.9, featuring a lightweight and optimized backend called HRX. This system enables high-performance integration with AMD's GPU, NPU, and CPU products, overcoming the limitations of existing ROCm backends on client systems. The HRX backend is a focused subset of ROCm, tailored for client operating systems and designed to maximize the performance of AMD's GPUs and NPUs. AMD's collaboration with the llama.cpp project has led to the development of this backend, which is now being considered for inclusion in ROCm. The initial debut of the HRX backend focuses on a minimal kernel library to run the unsloth/Qwen3-30B-A3B-Instruct-2507-GGUF Q4_K_M model efficiently on RDNA3 and RDNA3.5 architectures. The developers plan to expand the branch to support a wider range of models, operating systems, and systems in the future.",
  "summary": "Lemonade 11.9 is out today as the newest feature release to this AMD-backed, open-source local AI server solution across Linux, Windows, and macOS. Lemonade has long been focused on offering \"100% free and private\" AI use with local hardware whether it be GPUs, CPUs, or NPUs. With Lemonade 11.9's release today it's very interesting for having experimental ROCm HRX back-end support with Llama.cpp.…",
  "key_points": [
    "AMD releases Lemonade 11.9 AI server with HRX backend",
    "HRX backend optimizes AMD GPU, NPU, and CPU integration",
    "HRX backend developed for llama.cpp project and ROCm"
  ],
  "editors_take": "This development signals AMD's strategic push to enhance the performance and compatibility of its GPU, NPU, and CPU products with AI servers, potentially gaining an edge in the AI hardware market.",
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}