Skip to main content
POST
/
artifact
/
{artifact_type}
Create Artifact
curl --request POST \
  --url https://api.example.com/artifact/{artifact_type} \
  --header 'Content-Type: application/json' \
  --data '
{
  "url": "<string>",
  "name": "<string>",
  "download_url": "<string>"
}
'
{
  "metadata": {
    "metadata.name": "<string>",
    "metadata.id": "<string>",
    "metadata.type": "<string>"
  },
  "data": {
    "data.url": "<string>",
    "data.download_url": "<string>"
  }
}

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/GingerlyData247/SOTeam4-P2/llms.txt

Use this file to discover all available pages before exploring further.

​
Endpoint

​
artifact_type
string
required
Type of artifact to create. Must be one of:
  • model - Machine learning model (triggers HuggingFace ingestion)
  • dataset - Training or evaluation dataset
  • code - Code repository or training script

​
Authentication

Authentication headers may be required depending on deployment configuration.

​
Request Body

​
url
string
required
Source URL of the artifact. Maximum length: 2048 characters.For models: HuggingFace model URL or ID (e.g., https://huggingface.co/google-bert/bert-base-uncased or google-bert/bert-base-uncased)For datasets/code: Any valid URL pointing to the resource
​
name
string
Custom name for the artifact. If not provided, the name will be extracted from the URL path.
​
download_url
string
Optional pre-signed download URL for the artifact. If not provided, one may be automatically generated for model artifacts.

​
Response

Returns HTTP 201 on success with the created artifact details.
​
metadata
object
required
Artifact metadata
​
metadata.name
string
Name of the artifact
​
metadata.id
string
Unique identifier assigned to the artifact
​
metadata.type
string
Type of artifact: model, dataset, or code
​
data
object
required
Artifact data and URLs
​
data.url
string
Source URL of the artifact
​
data.download_url
string
Pre-signed download URL (if available)

​
Error Codes

  • 400 - Invalid artifact type or malformed request
  • 413 - URL exceeds maximum length (2048 characters)
  • 424 - Model ingestion rejected due to insufficient reviewedness score (< 0.5)
  • 500 - Internal server error during artifact creation

​
Model Artifact Behavior

When creating a model artifact, the system automatically:
  1. Normalizes the HuggingFace model ID from the URL
  2. Fetches model metadata and license information from HuggingFace
  3. Computes trustworthiness metrics (reviewedness, reproducibility, etc.)
  4. Validates reviewedness score - rejects models with reviewedness < 0.5
  5. Extracts parent dependencies for lineage tracking
  6. Creates a minimal artifact ZIP containing source URL
  7. Uploads the artifact to S3 storage
  8. Generates a pre-signed download URL

​
Dataset and Code Artifacts

For dataset and code artifacts:
  • No automatic metadata extraction
  • No metric computation
  • Simple registration with provided URL and name
  • No automatic storage upload (unless download_url is provided)

​
Examples

curl -X POST https://api.example.com/artifact/model \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://huggingface.co/google-bert/bert-base-uncased",
    "name": "BERT Base Uncased"
  }'

​
Response Example

{
  "metadata": {
    "name": "BERT Base Uncased",
    "id": "12345",
    "type": "model"
  },
  "data": {
    "url": "https://huggingface.co/google-bert/bert-base-uncased",
    "download_url": "https://s3.amazonaws.com/bucket/artifacts/model/12345.zip?signature=..."
  }
}

​
See Also

Build docs developers (and LLMs) love

Get started for freeTalk to us