PyTorch Lightning
PyPIpytorch-lightningPyTorch Lightning is a high-level training framework for PyTorch that handles distributed training, mixed-precision, checkpointing, and experiment logging. It is the standard abstraction layer for production ML training pipelines. Most large model fine-tuning workflows run through Lightning or a derivative (Lightning Fabric, Lit-LLM).
Checking PyTorch Lightning
pytorch-lightning 2.4.0 is a clean version with no known supply chain compromise. The response returns compromised: false with an empty sources array.
curl "https://api.attestd.io/v1/check?product=pytorch-lightning&version=2.4.0" \
-H "Authorization: Bearer YOUR_API_KEY"{
"product": "pytorch-lightning",
"version": "2.4.0",
"supported": true,
"risk_state": "none",
"supply_chain": {
"compromised": false,
"sources": [],
"malware_type": null,
"description": null,
"advisory_url": null,
"compromised_at": null,
"removed_at": null
},
"last_updated": "2026-05-01T00:00:00Z"
}Why this package is monitored
Training framework packages execute during model training runs that often process proprietary datasets, generate model checkpoints, and push artifacts to cloud storage. A compromised training library can exfiltrate dataset samples or embed backdoors into the model weights it produces.
Attestd monitors pytorch-lightning using the following detection sources:
registryManually curated advisories in the Attestd registry, verified by a human analyst. Confidence 1.0.
osvOSV.dev malicious-package advisories with IDs prefixed MAL-. Confidence 0.95.
pypi_yankVersions yanked on PyPI with a security-related yanked_reason annotation. Confidence 0.80.