{
  "id": "roots",
  "name": "ROOTS corpus",
  "builder": "BigScience",
  "release_date": {
    "value": "2023-03-07",
    "source": "https://arxiv.org/abs/2303.03915",
    "status": "partial",
    "note": "arXiv v1 date of the corpus paper; the corpus was used for BLOOM in 2022."
  },
  "availability": {
    "value": "gated",
    "checked": "2026-09-24",
    "source": "https://huggingface.co/datasets/bigscience-data/roots_en_wikipedia",
    "note": "Released as per-source subsets under bigscience-data/; one subset checked (gate: automatic approval)."
  },
  "content": {
    "value": "1.6TB of text in 59 languages (46 natural, 13 programming).",
    "source": "https://arxiv.org/abs/2303.03915",
    "status": "recorded"
  },
  "primary_sources": [
    "https://arxiv.org/abs/2303.03915",
    "https://arxiv.org/abs/2211.05100"
  ],
  "record_history": [
    {
      "date": "2026-09-24",
      "change": "created from primary sources (dataset records ruling)",
      "by": "wilson-pruitt + claude"
    }
  ]
}