{
  "id": 3774856,
  "title": "My local AI model was wasting VRAM by default, and changing one setting doubled its speed",
  "url": "https://urgent.news/2026/08/27/my-local-ai-model-was-wasting-vram-by-default-and-changing-one",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-27T16:30:15.000Z",
  "source": {
    "name": "XDA Developers",
    "slug": "xda-developers",
    "url": "https://www.xda-developers.com/my-local-ai-model-was-wasting-vram-by-default-and-lowering-one-setting-doubled-its-speed/"
  },
  "original_language": "en",
  "account": null,
  "summary": "When it comes to local AI, context matters for more than one reason",
  "key_points": [],
  "editors_take": null,
  "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."
}