Skip to main content
DELETE
/
artifacts
/
{artifact_type}
/
{id}
Delete Artifact
curl --request DELETE \
  --url https://api.example.com/artifacts/{artifact_type}/{id}
{
  "status": "<string>",
  "id": "<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

DELETE /artifacts/{artifact_type}/{id}

​
Parameters

​
artifact_type
string
required
Type of artifact to delete:
  • model - Machine learning model
  • dataset - Training or evaluation dataset
  • code - Code repository or script
​
id
string
required
Unique identifier of the artifact to delete.

​
Authentication

Authentication headers may be required depending on deployment configuration.

​
Response

Returns a confirmation message on successful deletion.
​
status
string
Deletion status - always "deleted" on success
​
id
string
ID of the deleted artifact

​
Error Codes

  • 404 - Artifact not found (already deleted or never existed)
  • 500 - Internal server error during deletion

​
Cleanup Behavior

When an artifact is deleted:

​
1. Registry Entry Removed

The artifact is removed from the registry’s in-memory model list (_registry._models).

​
2. Metadata Deleted

All associated metadata is removed, including:
  • Computed trust metrics (for models)
  • Lineage information
  • Tags and version information
  • Source URLs and download URLs

​
3. Storage Artifacts

Important: The current implementation does NOT automatically delete artifact files from S3 storage.Files remain at:
  • artifacts/model/{id}.zip
  • artifacts/dataset/{id}.zip
  • artifacts/code/{id}.zip
This may lead to orphaned storage objects. Consider implementing cleanup policies or manual deletion.

​
4. Lineage Impact

Deleting an artifact that is referenced as a parent in other artifacts’ lineage graphs may result in:
  • Broken lineage references
  • Incomplete dependency trees
  • Potential errors when querying /artifact/model/{id}/lineage for dependent artifacts
Consider checking lineage dependencies before deletion.

​
Idempotency

The delete endpoint is not idempotent:
  • First delete: Returns 200 with success message
  • Subsequent deletes: Returns 404 (artifact not found)

​
Examples

curl -X DELETE https://api.example.com/artifacts/model/12345

​
Response Examples

Success: 
{
  "status": "deleted",
  "id": "12345"
}
Not Found: 
{
  "detail": "Artifact does not exist."
}

​
Use Cases

​
Remove Outdated Models

Delete old model versions that are no longer needed: 
curl -X DELETE https://api.example.com/artifacts/model/old-version-123

​
Clean Up Failed Ingestions

If a model ingestion partially succeeded but has invalid metrics, delete and retry: 
# Delete the failed artifact
curl -X DELETE https://api.example.com/artifacts/model/failed-456

# Re-create with corrected data
curl -X POST https://api.example.com/artifact/model \
  -H "Content-Type: application/json" \
  -d '{"url": "https://huggingface.co/correct-model"}'

​
Remove Test Artifacts

Clean up artifacts created during testing: 
curl -X DELETE https://api.example.com/artifacts/dataset/test-dataset-1
curl -X DELETE https://api.example.com/artifacts/code/test-code-1

​
Best Practices

​
1. Check Dependencies First

Before deleting a model, verify it’s not used as a parent in lineage graphs: 
# Check if artifact has dependents
curl https://api.example.com/artifact/model/12345/lineage

# Look for edges where this artifact is the parent

​
2. Verify Deletion

Confirm the artifact is gone by attempting retrieval: 
# Should return 404
curl https://api.example.com/artifacts/model/12345

​
3. Manual Storage Cleanup

If you need to free S3 storage, manually delete artifact files: 
# Using AWS CLI
aws s3 rm s3://your-bucket/artifacts/model/12345.zip

​
4. Bulk Deletion

For deleting multiple artifacts, use a script to avoid manual repetition: 
#!/bin/bash
for id in 101 102 103 104; do
  curl -X DELETE https://api.example.com/artifacts/model/$id
done

​
System Reset

To delete all artifacts at once, use the system reset endpoint: 
curl -X DELETE https://api.example.com/reset
Destructive Operation: This resets the entire registry to default state, deleting all artifacts and metrics.

​
Implementation Details

The deletion logic is implemented in /home/daytona/workspace/source/src/api/routers/models.py:1207: 
@router.delete("/artifacts/{artifact_type}/{id}")
def artifact_delete(artifact_type: str, id: str):
    logger.info("DELETE /artifacts/%s/%s", artifact_type, id)
    
    # Attempt deletion from registry
    ok = _registry.delete(id)
    
    if not ok:
        # Artifact not found
        raise HTTPException(
            status_code=404, 
            detail="Artifact does not exist."
        )
    
    logger.info(
        "artifact_delete: deleted artifact_type=%s id=%s", 
        artifact_type, 
        id
    )
    
    return {"status": "deleted", "id": id}
The registry’s delete() method removes the artifact from the in-memory model list but does not perform S3 cleanup.

​
See Also

Build docs developers (and LLMs) love

Get started for freeTalk to us