{
  "id": "falcon-40b",
  "name": "falcon-40b",
  "identifiers": {
    "huggingface": [
      "tiiuae/falcon-40b"
    ]
  },
  "developer": "Technology Innovation Institute (TII)",
  "release_date": {
    "value": "2023-05-24",
    "source": "https://huggingface.co/api/models/tiiuae/falcon-40b",
    "status": "partial",
    "note": "HF repo creation date; may predate or postdate public release."
  },
  "weights_status": "open",
  "availability": {
    "value": "available",
    "checked": "2026-09-24",
    "source": "https://huggingface.co/tiiuae/falcon-40b"
  },
  "license": {
    "value": "apache-2.0",
    "source": "https://huggingface.co/tiiuae/falcon-40b/blob/05ab2ee8d6b593bdbab17d728de5c028a7a94d83/README.md",
    "status": "partial",
    "note": "From HF card metadata (uploader-declared); confirm against the license text."
  },
  "architecture": {
    "family": "decoder_only",
    "note": "family read from config 'architectures': ['FalconForCausalLM'].",
    "n_layers": {
      "value": 60,
      "source": "https://huggingface.co/tiiuae/falcon-40b/raw/05ab2ee8d6b593bdbab17d728de5c028a7a94d83/config.json",
      "status": "recorded"
    },
    "hidden_size": {
      "value": 8192,
      "source": "https://huggingface.co/tiiuae/falcon-40b/raw/05ab2ee8d6b593bdbab17d728de5c028a7a94d83/config.json",
      "status": "recorded"
    },
    "n_heads": {
      "value": 128,
      "source": "https://huggingface.co/tiiuae/falcon-40b/raw/05ab2ee8d6b593bdbab17d728de5c028a7a94d83/config.json",
      "status": "recorded"
    },
    "vocab_size": {
      "value": 65024,
      "source": "https://huggingface.co/tiiuae/falcon-40b/raw/05ab2ee8d6b593bdbab17d728de5c028a7a94d83/config.json",
      "status": "recorded"
    },
    "context_length": {
      "value": null,
      "status": "not_recorded"
    },
    "positional_encoding": {
      "value": "rotary (RoPE)",
      "source": "https://huggingface.co/tiiuae/falcon-40b",
      "status": "recorded",
      "note": "Same design as falcon-7b (card: 'broadly adapted from the GPT-3 paper')."
    }
  },
  "training_data": {
    "value": "1,500B tokens: RefinedWeb (filtered, deduplicated web crawl) plus curated corpora. Same recipe as the 7B model.",
    "source": "https://huggingface.co/tiiuae/falcon-40b",
    "status": "recorded"
  },
  "techniques": [],
  "primary_sources": [
    "https://huggingface.co/tiiuae/falcon-40b",
    "https://huggingface.co/tiiuae/falcon-40b/raw/05ab2ee8d6b593bdbab17d728de5c028a7a94d83/config.json"
  ],
  "record_history": [
    {
      "date": "2026-09-24",
      "change": "ingested as candidate from HF (tiiuae/falcon-40b@05ab2ee8d6b593bdbab17d728de5c028a7a94d83)",
      "by": "ingest_hf.py"
    },
    {
      "date": "2026-09-24",
      "change": "preparer: developer, positional encoding, trained_on accepted, design_follows proposed (needs confirmation)",
      "by": "claude (preparer, Sonnet 5)"
    },
    {
      "date": "2026-09-24",
      "change": "preparer: design_follows edge independently verified against the 40B card text",
      "by": "claude (preparer, Sonnet 5)"
    },
    {
      "date": "2026-09-24",
      "change": "preparer cleanup: removed 3 stale flag(s), filled training_data",
      "by": "claude (preparer, Sonnet 5)"
    },
    {
      "date": "2026-09-24",
      "change": "reviewed and promoted from staging (2 edge(s) accepted)",
      "by": "Wilson Pruitt"
    }
  ]
}