Hugging Face · Schema
Hugging Face Model
Schema for a machine learning model hosted on the Hugging Face Hub, including metadata, configuration, and repository information.
Properties
| Name | Type | Description |
|---|---|---|
| _id | string | Internal unique identifier for the model |
| id | string | Model repository ID in the format author/model-name or model-name |
| modelId | string | Alias for the model repository ID |
| author | string | Author or organization that owns the model |
| sha | string | Latest Git commit SHA of the model repository |
| lastModified | string | Timestamp of the last modification to the repository |
| createdAt | string | Timestamp when the model repository was created |
| private | boolean | Whether the model repository is private |
| disabled | boolean | Whether the model has been disabled |
| gated | object | Access gating configuration. False means no gating, 'auto' or 'manual' indicates gated access. |
| pipeline_tag | string | The primary task/pipeline this model is designed for |
| tags | array | Tags associated with the model including library, language, license, and custom tags |
| library_name | string | Primary ML library used by the model |
| downloads | integer | Total number of downloads in the last 30 days |
| downloadsAllTime | integer | Total number of all-time downloads |
| likes | integer | Total number of likes/favorites |
| siblings | array | List of files in the model repository |
| spaces | array | List of Space IDs that use this model |
| safetensors | object | Safetensors metadata including parameter counts |
| config | object | Model configuration from config.json |
| cardData | object | Parsed metadata from the model card (README.md YAML front matter) |
| transformersInfo | object | Transformers library-specific information |
| widgetData | array | Widget example data for the model card inference widget |
JSON Schema
{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"$id": "https://huggingface.co/schemas/model.json",
"title": "Hugging Face Model",
"description": "Schema for a machine learning model hosted on the Hugging Face Hub, including metadata, configuration, and repository information.",
"type": "object",
"required": [
"id"
],
"properties": {
"_id": {
"type": "string",
"description": "Internal unique identifier for the model"
},
"id": {
"type": "string",
"description": "Model repository ID in the format author/model-name or model-name",
"examples": [
"bert-base-uncased",
"meta-llama/Llama-3-70b-chat-hf",
"openai/whisper-large-v3"
]
},
"modelId": {
"type": "string",
"description": "Alias for the model repository ID"
},
"author": {
"type": "string",
"description": "Author or organization that owns the model",
"examples": [
"meta-llama",
"google",
"microsoft"
]
},
"sha": {
"type": "string",
"description": "Latest Git commit SHA of the model repository",
"pattern": "^[0-9a-f]{40}$"
},
"lastModified": {
"type": "string",
"format": "date-time",
"description": "Timestamp of the last modification to the repository"
},
"createdAt": {
"type": "string",
"format": "date-time",
"description": "Timestamp when the model repository was created"
},
"private": {
"type": "boolean",
"description": "Whether the model repository is private",
"default": false
},
"disabled": {
"type": "boolean",
"description": "Whether the model has been disabled",
"default": false
},
"gated": {
"oneOf": [
{
"type": "boolean"
},
{
"type": "string",
"enum": [
"auto",
"manual"
]
}
],
"description": "Access gating configuration. False means no gating, 'auto' or 'manual' indicates gated access."
},
"pipeline_tag": {
"type": "string",
"description": "The primary task/pipeline this model is designed for",
"enum": [
"text-generation",
"text-classification",
"token-classification",
"question-answering",
"summarization",
"translation",
"fill-mask",
"text2text-generation",
"feature-extraction",
"sentence-similarity",
"zero-shot-classification",
"table-question-answering",
"conversational",
"image-classification",
"object-detection",
"image-segmentation",
"image-to-text",
"text-to-image",
"text-to-video",
"text-to-speech",
"text-to-audio",
"automatic-speech-recognition",
"audio-classification",
"image-text-to-text",
"visual-question-answering",
"document-question-answering",
"depth-estimation",
"image-to-image",
"reinforcement-learning",
"robotics",
"video-classification"
]
},
"tags": {
"type": "array",
"items": {
"type": "string"
},
"description": "Tags associated with the model including library, language, license, and custom tags",
"examples": [
[
"transformers",
"pytorch",
"en",
"text-generation",
"license:apache-2.0"
]
]
},
"library_name": {
"type": "string",
"description": "Primary ML library used by the model",
"enum": [
"transformers",
"diffusers",
"timm",
"sentence-transformers",
"spacy",
"allennlp",
"flair",
"asteroid",
"espnet",
"speechbrain",
"adapter-transformers",
"fastai",
"stable-baselines3",
"ml-agents",
"open_clip",
"peft",
"setfit",
"span-marker",
"keras",
"sklearn",
"onnx",
"safetensors",
"tensorboard"
]
},
"downloads": {
"type": "integer",
"description": "Total number of downloads in the last 30 days",
"minimum": 0
},
"downloadsAllTime": {
"type": "integer",
"description": "Total number of all-time downloads",
"minimum": 0
},
"likes": {
"type": "integer",
"description": "Total number of likes/favorites",
"minimum": 0
},
"siblings": {
"type": "array",
"items": {
"type": "object",
"properties": {
"rfilename": {
"type": "string",
"description": "Relative file path within the repository"
},
"size": {
"type": "integer",
"description": "File size in bytes"
},
"blobId": {
"type": "string",
"description": "Git blob ID"
},
"lfs": {
"type": "object",
"description": "Git LFS metadata",
"properties": {
"sha256": {
"type": "string",
"description": "SHA-256 hash of the LFS file"
},
"size": {
"type": "integer",
"description": "Actual file size in bytes"
},
"pointerSize": {
"type": "integer",
"description": "Size of the LFS pointer file"
}
}
}
}
},
"description": "List of files in the model repository"
},
"spaces": {
"type": "array",
"items": {
"type": "string"
},
"description": "List of Space IDs that use this model"
},
"safetensors": {
"type": "object",
"description": "Safetensors metadata including parameter counts",
"properties": {
"parameters": {
"type": "object",
"additionalProperties": {
"type": "integer"
},
"description": "Parameter counts by dtype (e.g., F16, BF16, F32)"
},
"total": {
"type": "integer",
"description": "Total parameter count"
}
}
},
"config": {
"type": "object",
"description": "Model configuration from config.json",
"properties": {
"architectures": {
"type": "array",
"items": {
"type": "string"
},
"description": "Model architecture classes"
},
"model_type": {
"type": "string",
"description": "Model type identifier (e.g., bert, gpt2, llama)"
},
"tokenizer_config": {
"type": "object",
"description": "Tokenizer configuration"
}
}
},
"cardData": {
"type": "object",
"description": "Parsed metadata from the model card (README.md YAML front matter)",
"properties": {
"language": {
"oneOf": [
{
"type": "string"
},
{
"type": "array",
"items": {
"type": "string"
}
}
],
"description": "Language(s) supported by the model"
},
"license": {
"type": "string",
"description": "License identifier",
"examples": [
"apache-2.0",
"mit",
"cc-by-4.0",
"llama3"
]
},
"library_name": {
"type": "string"
},
"tags": {
"type": "array",
"items": {
"type": "string"
}
},
"datasets": {
"type": "array",
"items": {
"type": "string"
},
"description": "Datasets used for training"
},
"metrics": {
"type": "array",
"items": {
"type": "string"
},
"description": "Evaluation metrics reported"
},
"base_model": {
"oneOf": [
{
"type": "string"
},
{
"type": "array",
"items": {
"type": "string"
}
}
],
"description": "Base model(s) this model is derived from"
},
"pipeline_tag": {
"type": "string"
},
"model-index": {
"type": "array",
"items": {
"type": "object"
},
"description": "Evaluation results in model index format"
}
}
},
"transformersInfo": {
"type": "object",
"description": "Transformers library-specific information",
"properties": {
"auto_model": {
"type": "string",
"description": "AutoModel class to use (e.g., AutoModelForCausalLM)"
},
"pipeline_tag": {
"type": "string",
"description": "Pipeline tag inferred by transformers"
},
"processor": {
"type": "string",
"description": "Processor class to use"
}
}
},
"widgetData": {
"type": "array",
"items": {
"type": "object"
},
"description": "Widget example data for the model card inference widget"
}
}
}