{
  "id": "microsoft-phi-4-multimodal-instruct",
  "name": "Phi-4-multimodal-instruct",
  "developer": "Microsoft",
  "family": "Phi-4-multimodal",
  "variant": "Phi-4-multimodal-instruct",
  "model": {
    "parameters": "5.57446B",
    "total_parameters_billions": 5.574460384,
    "context": {
      "value": 131072,
      "display": "131,072 tokens",
      "status": "verified",
      "evidence": [
        {
          "source": "config.json",
          "detail": "max_position_embeddings"
        }
      ]
    },
    "modality": "Audio → text",
    "tags": [
      "transformers",
      "safetensors",
      "phi4mm",
      "text-generation",
      "nlp",
      "code",
      "audio",
      "automatic-speech-recognition",
      "speech-summarization",
      "speech-translation",
      "visual-question-answering",
      "phi-4-multimodal",
      "phi",
      "phi-4-mini",
      "custom_code",
      "multilingual",
      "ar",
      "zh",
      "cs",
      "da"
    ]
  },
  "weights": {
    "status": "verified-available",
    "repository": "https://huggingface.co/microsoft/Phi-4-multimodal-instruct",
    "access": "public",
    "exact_files": [
      "model-00001-of-00003.safetensors",
      "model-00002-of-00003.safetensors",
      "model-00003-of-00003.safetensors",
      "model.safetensors.index.json",
      "speech-lora/adapter_model.safetensors",
      "vision-lora/adapter_model.safetensors"
    ],
    "file_count": 6,
    "formats": [
      "Safetensors"
    ],
    "precision_availability": {
      "bf16": {
        "available": true,
        "status": "verified"
      },
      "fp16": {
        "available": false,
        "status": "verified"
      },
      "fp8": {
        "available": false,
        "status": "verified"
      },
      "int8": {
        "available": false,
        "status": "verified"
      },
      "int4": {
        "available": false,
        "status": "verified"
      },
      "gguf": {
        "available": false,
        "status": "verified"
      }
    },
    "dtype_evidence": [
      "BF16",
      "BFLOAT16"
    ],
    "evidence": [
      {
        "source": "https://huggingface.co/api/models/microsoft/Phi-4-multimodal-instruct",
        "type": "Hugging Face model API siblings"
      }
    ]
  },
  "license": {
    "id": "mit",
    "name": "MIT",
    "url": "https://huggingface.co/microsoft/Phi-4-multimodal-instruct/resolve/main/LICENSE",
    "status": "verified",
    "commercial_use": {
      "status": "allowed",
      "label": "Commercial use allowed under MIT conditions",
      "basis": "https://huggingface.co/microsoft/Phi-4-multimodal-instruct/resolve/main/LICENSE"
    },
    "repository_license_files": [
      "LICENSE"
    ],
    "verification": {
      "status": "license-file-checked",
      "checked_at": "2026-10-03",
      "source": "https://huggingface.co/microsoft/Phi-4-multimodal-instruct/resolve/main/LICENSE"
    }
  },
  "hardware": {
    "status": "estimated",
    "weight_only_gb": {
      "bf16_fp16": 11.15,
      "fp8_int8": 5.57,
      "int4": 2.79
    },
    "note": "Weight-only decimal-GB estimate from total parameter count; excludes KV cache, activations, runtime overhead, optimizer state and sharding."
  },
  "lineage": {
    "family": "Phi-4-multimodal",
    "variant": "Phi-4-multimodal-instruct",
    "base_model": {
      "status": "not-declared",
      "models": [],
      "relation": null,
      "evidence": []
    }
  },
  "training_assets": {
    "training_recipe": {
      "status": "training-described-in-model-card",
      "files": [],
      "links": [
        "https://huggingface.co/microsoft/Phi-4-multimodal-instruct/blob/main/added_tokens.json",
        "https://huggingface.co/microsoft/Phi-4-multimodal-instruct/resolve/main/sample_finetune_speech.py",
        "https://huggingface.co/microsoft/Phi-4-multimodal-instruct/resolve/main/sample_finetune_vision.py",
        "https://huggingface.co/microsoft/Phi-4-multimodal-instruct#appendix-b-fine-tuning-korean-speech",
        "https://huggingface.co/daekeun-ml/Phi-4-multimodal-finetune-ko-speech",
        "https://huggingface.co/daekeun-ml/Phi-4-multimodal-finetune-ko-speech/tree/main/demos",
        "https://huggingface.co/microsoft/Phi-4-multimodal-instruct",
        "https://gist.github.com/seastar105/d1d8983b27611370528e3b194dcc5577#file-main-py",
        "https://colab.research.google.com/drive/1JAQdpX3BtIgDmTLlnHgstKfGw7HjSfej?usp=sharing"
      ],
      "evidence": [
        "README.md"
      ]
    },
    "data_disclosure": {
      "status": "described-in-model-card",
      "datasets": [],
      "evidence": [
        {
          "source": "README.md",
          "detail": "data/dataset section present"
        }
      ]
    },
    "evaluations": "Not classified by automated verifier",
    "intermediate_checkpoints": "Not classified by automated verifier"
  },
  "sources": [
    {
      "type": "model_repository",
      "url": "https://huggingface.co/microsoft/Phi-4-multimodal-instruct",
      "tier": "primary"
    },
    {
      "type": "model_api",
      "url": "https://huggingface.co/api/models/microsoft/Phi-4-multimodal-instruct",
      "tier": "primary"
    },
    {
      "type": "model_card",
      "url": "https://huggingface.co/microsoft/Phi-4-multimodal-instruct/blob/main/README.md",
      "tier": "primary",
      "available": true
    },
    {
      "type": "config",
      "url": "https://huggingface.co/microsoft/Phi-4-multimodal-instruct/blob/main/config.json",
      "tier": "primary",
      "available": true
    },
    {
      "type": "license",
      "url": "https://huggingface.co/microsoft/Phi-4-multimodal-instruct/resolve/main/LICENSE",
      "tier": "primary-or-declared"
    }
  ],
  "verification": {
    "level": "field-verified",
    "checked_at": "2026-10-03",
    "checked_at_iso": "2026-10-03T10:18:29+00:00",
    "method": "Hugging Face API + repository file list + model card + config.json",
    "mode": "repository-revision-unchanged",
    "full_verified_at": "2026-10-03",
    "readme_accessible": true,
    "config_accessible": true,
    "fields": {
      "license": "verified",
      "commercial_use": "classified",
      "context": "verified",
      "weight_files": "verified",
      "formats": "verified",
      "base_model": "not-declared",
      "training_recipe": "training-described-in-model-card",
      "data_disclosure": "described-in-model-card",
      "runtime_support": "source-derived-not-runtime-tested"
    },
    "history": [
      {
        "date": "2026-10-03",
        "event": "Repository revision check",
        "detail": "Official Hugging Face repository SHA unchanged; prior field evidence retained."
      },
      {
        "date": "2026-10-03",
        "event": "Field-by-field verification",
        "detail": "Checked Hugging Face API metadata, exact repository file list, model card and config when accessible."
      },
      {
        "date": "2026-10-03",
        "event": "Registry record created",
        "detail": "Previous verification level: discovered-pending."
      }
    ]
  },
  "discovery": {
    "source": "Hugging Face public model index",
    "repo_id": "microsoft/Phi-4-multimodal-instruct",
    "rank": 717,
    "downloads": 211100,
    "likes": 1615,
    "last_modified": "2025-12-10T20:18:10+00:00",
    "created_at": "2025-02-24T22:33:32+00:00",
    "library_name": "transformers",
    "pipeline_tag": "automatic-speech-recognition",
    "discovered_at": "2026-10-03T10:18:27+00:00",
    "pinned": false
  },
  "runtime_support": {
    "declared_library": "Transformers",
    "pipeline_tag": "automatic-speech-recognition",
    "declared": [
      "Transformers"
    ],
    "mentioned_in_model_card": [
      "ONNX Runtime",
      "vLLM"
    ],
    "artifact_signals": [],
    "tested_by_openmodelweights": false
  },
  "hub": {
    "last_modified": "2025-12-10T20:18:10.000Z",
    "created_at": "2025-02-24T22:33:32.000Z",
    "downloads": 211100,
    "likes": 1615,
    "sha": "93f923e1a7727d1c4f446756212d9d3e8fcc5d81",
    "library_name": "transformers",
    "pipeline_tag": "automatic-speech-recognition",
    "tags": [
      "transformers",
      "safetensors",
      "phi4mm",
      "text-generation",
      "nlp",
      "code",
      "audio",
      "automatic-speech-recognition",
      "speech-summarization",
      "speech-translation",
      "visual-question-answering",
      "phi-4-multimodal",
      "phi",
      "phi-4-mini",
      "custom_code",
      "multilingual",
      "ar",
      "zh",
      "cs",
      "da",
      "nl",
      "en",
      "fi",
      "fr",
      "de",
      "he",
      "hu",
      "it",
      "ja",
      "ko",
      "no",
      "pl",
      "pt",
      "ru",
      "es",
      "sv",
      "th",
      "tr",
      "uk",
      "arxiv:2503.01743",
      "arxiv:2407.13833",
      "license:mit",
      "eval-results",
      "region:us"
    ]
  }
}
