{
  "id": "bloom-7b1",
  "name": "BLOOM 7B1",
  "identifiers": {
    "huggingface": [
      "bigscience/bloom-7b1"
    ]
  },
  "developer": "BigScience",
  "release_date": {
    "value": "2022-07-11",
    "source": "https://huggingface.co/bigscience/bloom-7b1/blob/6232703e399354503377bf59dfbb8397fd569e4a/README.md",
    "status": "partial",
    "note": "Card: 'Release Date Estimate: Monday, 11.July.2022'."
  },
  "weights_status": "open",
  "availability": {
    "value": "available",
    "checked": "2026-09-24",
    "source": "https://huggingface.co/bigscience/bloom-7b1"
  },
  "license": {
    "value": "BigScience RAIL License v1.0 (open access with use restrictions)",
    "source": "https://huggingface.co/bigscience/bloom-7b1/blob/6232703e399354503377bf59dfbb8397fd569e4a/README.md",
    "status": "recorded"
  },
  "architecture": {
    "family": "decoder_only",
    "note": "Card: 'Modified from Megatron-LM GPT2' (the Megatron codebase, not OpenAI's weights). Paper: ALiBi instead of position embeddings, plus a LayerNorm on the embedding layer. GeLU activation.",
    "n_layers": {
      "value": 30,
      "source": "https://huggingface.co/bigscience/bloom-7b1/blob/6232703e399354503377bf59dfbb8397fd569e4a/README.md",
      "status": "recorded"
    },
    "hidden_size": {
      "value": 4096,
      "source": "https://huggingface.co/bigscience/bloom-7b1/blob/6232703e399354503377bf59dfbb8397fd569e4a/README.md",
      "status": "recorded"
    },
    "n_heads": {
      "value": 32,
      "source": "https://huggingface.co/bigscience/bloom-7b1/blob/6232703e399354503377bf59dfbb8397fd569e4a/README.md",
      "status": "recorded"
    },
    "vocab_size": {
      "value": 250680,
      "source": "https://huggingface.co/bigscience/bloom-7b1/blob/6232703e399354503377bf59dfbb8397fd569e4a/README.md",
      "status": "recorded",
      "note": "Config vocab_size is 250880 (padded)."
    },
    "positional_encoding": {
      "value": "ALiBi (attention bias by distance; no position embeddings)",
      "source": "https://arxiv.org/abs/2211.05100",
      "status": "recorded"
    },
    "context_length": {
      "value": 2048,
      "source": "https://huggingface.co/bigscience/bloom-7b1/blob/6232703e399354503377bf59dfbb8397fd569e4a/README.md",
      "status": "recorded"
    }
  },
  "training_data": {
    "value": "ROOTS corpus: 46 natural and 13 programming languages; 1.5TB of text, 350B unique tokens.",
    "source": "https://arxiv.org/abs/2211.05100",
    "status": "recorded"
  },
  "techniques": [
    "transformer-decoder",
    "alibi"
  ],
  "primary_sources": [
    "https://arxiv.org/abs/2211.05100",
    "https://huggingface.co/bigscience/bloom-7b1/blob/6232703e399354503377bf59dfbb8397fd569e4a/README.md",
    "https://huggingface.co/bigscience/bloom-7b1/raw/6232703e399354503377bf59dfbb8397fd569e4a/config.json"
  ],
  "record_history": [
    {
      "date": "2026-09-24",
      "change": "created from primary sources (Phase 1 seed, batch 2)",
      "by": "wilson-pruitt + claude"
    },
    {
      "date": "2026-09-24",
      "change": "availability checked and recorded",
      "by": "wilson-pruitt + claude"
    }
  ]
}