{
  "id": "the-stack-dedup",
  "name": "The Stack (near-deduplicated)",
  "identifiers": {
    "huggingface": [
      "bigcode/the-stack-dedup"
    ]
  },
  "builder": "BigCode",
  "release_date": {
    "value": "2022-10-06",
    "source": "https://huggingface.co/api/datasets/bigcode/the-stack-dedup",
    "status": "partial",
    "note": "HF dataset repo creation date, not the builder's own release announcement."
  },
  "availability": {
    "value": "gated",
    "checked": "2026-09-24",
    "source": "https://huggingface.co/datasets/bigcode/the-stack-dedup",
    "note": "HF API reports gated: auto (approval step)."
  },
  "content": {
    "value": "StarCoder paper: The Stack is 'a publicly available pre-training dataset for Code LLMs with a transparent data governance framework ... 6.4 TB of permissively licensed source code in 384 programming languages'; StarCoder used v1.2 after near-deduplication. Whether the HF repo the-stack-dedup is exactly the v1.2 dedup used is not checked (card gated).",
    "source": "https://arxiv.org/abs/2305.06161",
    "status": "partial"
  },
  "primary_sources": [
    "https://arxiv.org/abs/2305.06161",
    "https://arxiv.org/abs/2211.15533",
    "https://huggingface.co/datasets/bigcode/the-stack-dedup"
  ],
  "record_history": [
    {
      "date": "2026-09-24",
      "change": "created from primary sources (tranche 2 dataset records)",
      "by": "wilson-pruitt + claude"
    }
  ]
}