{
  "id": 6305946,
  "title": "The Flag We Tuned Around Got Deleted",
  "url": "https://urgent.news/2026/09/08/the-flag-we-tuned-around-got-deleted",
  "topic": "tech",
  "section": "Tech",
  "published": "2026-09-08T18:06:52.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/devbrewery/the-flag-we-tuned-around-got-deleted-31ge"
  },
  "original_language": "en",
  "account": "For a dual GPU Tesla P40 setup, the most crucial llama.cpp flag was \"-sm row\". This flag divided every layer's tensors between both GPUs, resulting in nearly twice the throughput compared to the alternative \"layer split\". Every model stack I built was optimized around \"-sm row\". However, in July 2026, the upstream llama.cpp project deleted the flag entirely, deleting a significant performance rule that had been in place twice. This story covers five acts: Act 1 - Row wins in February 2026, Act 2 - Regression incident in March 2026, Act 3 - Fast mode becomes the wrong mode in May 2026, Act 4 - Upstream deletes row in July 2026, and Act 5 - The win comes back from elsewhere in the present.",
  "summary": "The single most important llama.cpp flag for my dual Tesla P40 setup was -sm row . It split every layer's tensors across both GPUs and it was worth nearly double the throughput of the alternative: 12-14 tokens/sec against about 7 for layer split. Every stack I built was tuned around it. In July 2026, upstream llama.cpp deleted it. Not deprecated. Deleted. This is the story of a performance rule…",
  "key_points": [
    "-sm row flag optimized dual GPU throughput",
    "Flag deleted in July 2026, disrupting performance",
    "Win restored through alternative configurations"
  ],
  "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."
}