{
  "id": "mixtral-8x7b-v0-1",
  "name": "Mixtral-8x7B-v0.1",
  "identifiers": {
    "huggingface": [
      "mistralai/Mixtral-8x7B-v0.1"
    ]
  },
  "developer": "Mistral AI",
  "release_date": {
    "value": "2023-12-11",
    "source": "https://mistral.ai/news/mixtral-of-experts",
    "status": "recorded"
  },
  "weights_status": "open",
  "availability": {
    "value": "available",
    "checked": "2026-09-24",
    "source": "https://huggingface.co/mistralai/Mixtral-8x7B-v0.1"
  },
  "license": {
    "value": "Apache 2.0",
    "source": "https://mistral.ai/news/mixtral-of-experts",
    "status": "recorded"
  },
  "architecture": {
    "family": "decoder_only_moe",
    "note": "Sparse mixture-of-experts. Paper: 'Mixtral has the same architecture as Mistral 7B, with the difference that each layer is composed of 8 feedforward blocks (i.e. experts).' A router selects 2 of 8 experts per token per layer. 46.7B total parameters, 12.9B active per token. Pretrained from scratch, not fine-tuned from Mistral 7B weights (Mistral AI release post).",
    "n_experts": {
      "value": 8,
      "source": "https://arxiv.org/abs/2310.06825",
      "status": "recorded"
    },
    "n_experts_per_tok": {
      "value": 2,
      "source": "https://arxiv.org/abs/2310.06825",
      "status": "recorded"
    },
    "n_layers": {
      "value": null,
      "status": "not_recorded"
    },
    "hidden_size": {
      "value": null,
      "status": "not_recorded"
    },
    "n_heads": {
      "value": null,
      "status": "not_recorded"
    },
    "vocab_size": {
      "value": null,
      "status": "not_recorded"
    },
    "positional_encoding": {
      "value": null,
      "status": "not_recorded",
      "note": "Config not fetched this session; likely rotary as in Mistral 7B, per the 'same architecture' claim, but not independently confirmed."
    }
  },
  "training_data": {
    "value": "Pretrained from scratch; neither the paper nor the release post discloses the training corpus.",
    "source": "https://mistral.ai/news/mixtral-of-experts",
    "status": "recorded"
  },
  "techniques": [],
  "primary_sources": [
    "https://huggingface.co/mistralai/Mixtral-8x7B-v0.1",
    "https://huggingface.co/mistralai/Mixtral-8x7B-v0.1/raw/fc7ac94680e38d7348cfa806e51218e6273104b0/config.json"
  ],
  "record_history": [
    {
      "date": "2026-09-24",
      "change": "ingested as candidate from HF (mistralai/Mixtral-8x7B-v0.1@fc7ac94680e38d7348cfa806e51218e6273104b0)",
      "by": "ingest_hf.py"
    },
    {
      "date": "2026-09-24",
      "change": "preparer: developer, architecture (MoE, from paper), license, release date, same_architecture_retrained to mistral-7b-v0-1 added",
      "by": "claude (preparer, Sonnet 5)"
    },
    {
      "date": "2026-09-24",
      "change": "preparer cleanup: removed 2 stale flag(s), filled training_data",
      "by": "claude (preparer, Sonnet 5)"
    },
    {
      "date": "2026-09-24",
      "change": "reviewed and promoted from staging (1 edge(s) accepted)",
      "by": "Wilson Pruitt"
    }
  ]
}