{
  "id": 11004832,
  "title": "Cohere’s faster query model barely dents retrieval quality in its tests",
  "url": "https://urgent.news/2026/09/30/coheres-faster-query-model-barely-dents-retrieval-quality-in-its-tests",
  "topic": "tech",
  "section": "Tech",
  "published": "2026-09-30T18:41:53.000Z",
  "source": {
    "name": "The New Stack",
    "slug": "the-new-stack",
    "url": "https://thenewstack.io/cohere-embed-pro-fast/"
  },
  "original_language": "en",
  "account": null,
  "summary": "Cohere has released Embed 5, a faster and cheaper alternative to its Embed 5 Pro model for retrieval-augmented generation (RAG) workloads. While the faster model, Embed 5 Fast, shows a slight decrease in retrieval quality, Cohere's tests indicate that the tradeoff is minimal. Both models share the same embedding space, allowing teams to switch between them without re-embedding their corpus. Embed 5 supports various data types, including text, images, and fused documents, and offers storage savings with different vector dimensions and formats.",
  "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."
}