{
  "id": "falcon-7b",
  "name": "Falcon 7B",
  "identifiers": {
    "huggingface": [
      "tiiuae/falcon-7b"
    ]
  },
  "developer": "Technology Innovation Institute (TII)",
  "release_date": {
    "value": "2023-04-24",
    "source": "https://huggingface.co/api/models/tiiuae/falcon-7b",
    "status": "partial",
    "note": "Creation date of the developer's HF repository (may predate public release). Card: training happened in early March 2023."
  },
  "weights_status": "open",
  "availability": {
    "value": "available",
    "checked": "2026-09-24",
    "source": "https://huggingface.co/tiiuae/falcon-7b"
  },
  "license": {
    "value": "Apache 2.0",
    "source": "https://huggingface.co/tiiuae/falcon-7b/blob/ec89142b67d748a1865ea4451372db8313ada0d8/README.md",
    "status": "recorded"
  },
  "architecture": {
    "family": "decoder_only",
    "note": "Card: 'broadly adapted from the GPT-3 paper' with rotary embeddings, multiquery attention + FlashAttention, and parallel attention/MLP with a single layer norm. d_model raised to 4544 'to compensate for multiquery'; head_dim 64.",
    "n_layers": {
      "value": 32,
      "source": "https://huggingface.co/tiiuae/falcon-7b/blob/ec89142b67d748a1865ea4451372db8313ada0d8/README.md",
      "status": "recorded"
    },
    "hidden_size": {
      "value": 4544,
      "source": "https://huggingface.co/tiiuae/falcon-7b/blob/ec89142b67d748a1865ea4451372db8313ada0d8/README.md",
      "status": "recorded"
    },
    "n_heads": {
      "value": 71,
      "source": "https://huggingface.co/tiiuae/falcon-7b/raw/ec89142b67d748a1865ea4451372db8313ada0d8/config.json",
      "status": "recorded"
    },
    "n_kv_heads": {
      "value": 1,
      "source": "https://huggingface.co/tiiuae/falcon-7b/raw/ec89142b67d748a1865ea4451372db8313ada0d8/config.json",
      "status": "recorded",
      "note": "Config multi_query: true (one shared key/value head)."
    },
    "vocab_size": {
      "value": 65024,
      "source": "https://huggingface.co/tiiuae/falcon-7b/blob/ec89142b67d748a1865ea4451372db8313ada0d8/README.md",
      "status": "recorded"
    },
    "positional_encoding": {
      "value": "rotary (RoPE)",
      "source": "https://huggingface.co/tiiuae/falcon-7b/blob/ec89142b67d748a1865ea4451372db8313ada0d8/README.md",
      "status": "recorded"
    },
    "context_length": {
      "value": 2048,
      "source": "https://huggingface.co/tiiuae/falcon-7b/blob/ec89142b67d748a1865ea4451372db8313ada0d8/README.md",
      "status": "recorded"
    }
  },
  "training_data": {
    "value": "1,500B tokens: RefinedWeb (filtered, deduplicated web crawl; 79% English) plus curated corpora, some 'inspired by The Pile'.",
    "source": "https://huggingface.co/tiiuae/falcon-7b/blob/ec89142b67d748a1865ea4451372db8313ada0d8/README.md",
    "status": "recorded"
  },
  "techniques": [
    "transformer-decoder",
    "rotary-position-embedding",
    "multi-query-attention",
    "parallel-attention-ffn"
  ],
  "primary_sources": [
    "https://huggingface.co/tiiuae/falcon-7b/blob/ec89142b67d748a1865ea4451372db8313ada0d8/README.md",
    "https://arxiv.org/abs/2311.16867",
    "https://huggingface.co/tiiuae/falcon-7b/raw/ec89142b67d748a1865ea4451372db8313ada0d8/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"
    }
  ]
}