Ctrl + K
DocumentationLog inGet started

tessl/pypi-rembg

tessl install tessl/pypi-rembg@2.0.0

Remove image background using advanced AI models including U-Net, BiRefNet, and SAM with support for multiple input formats and GPU acceleration

Agent Success

Agent success rate when using this tile

84%

Improvement

Agent success rate improvement when using this tile compared to baseline

0.94x

Baseline

Agent success rate without this tile

89%

rubric.jsonevals/scenario-1/

{
  "context": "This criteria evaluates how well the engineer uses rembg's model management API to implement downloading, listing, and checking model availability. The focus is on proper use of rembg's built-in functions rather than implementing custom model management logic.",
  "type": "weighted_checklist",
  "checklist": [
    {
      "name": "Uses download_models function",
      "description": "Correctly imports and uses the rembg.download_models() function to download models programmatically",
      "max_score": 30
    },
    {
      "name": "Uses sessions_names",
      "description": "Imports and uses rembg.sessions.sessions_names or rembg.sessions.sessions to list available or downloaded models",
      "max_score": 25
    },
    {
      "name": "Model existence verification",
      "description": "Implements model existence checking using rembg.sessions.sessions dictionary or by attempting to create a session with new_session()",
      "max_score": 25
    },
    {
      "name": "Error handling",
      "description": "Properly handles invalid model names and returns appropriate boolean values as specified in the API",
      "max_score": 10
    },
    {
      "name": "Correct return types",
      "description": "All functions return the correct types as specified: bool for download_models and is_model_downloaded, list[str] for list_downloaded_models",
      "max_score": 10
    }
  ]
}

Version

Workspace
tessl
Visibility
Public
Created
Last updated
Describes
pypipkg:pypi/rembg@2.0.x
tile.json