CVE-2026-71281

A flaw was found in peft. The LoRA-GA and CorDA initialization modules within Hugging Face peft improperly call torch.load without the weightsonly=True parameter. This oversight allows for full pickle deserialization when loading a malicious cache or covariance file. A remote attacker could explo...

CVE-2026-71281

Hugging Face peft's LoRA-GA and CorDA initialization modules src/peft/tuners/lora/corda.py lines 102 and 163, and src/peft/tuners/lora/loraga.py line 101 call torch.load on config-specified cache/covariance files without weightsonly=True, bypassing peft's own safe-loading wrapper used elsewhere i...

CVE-2026-71281 peft Unsafe Deserialization via torch.load() Without weights_only in LoRA-GA and CorDA Modules

Hugging Face peft's LoRA-GA and CorDA initialization modules src/peft/tuners/lora/corda.py lines 102 and 163, and src/peft/tuners/lora/loraga.py line 101 call torch.load on config-specified cache/covariance files without weightsonly=True, bypassing peft's own safe-loading wrapper used elsewhere i...

CVE-2026-71281

Hugging Face peft's LoRA-GA and CorDA initialization modules src/peft/tuners/lora/corda.py lines 102 and 163, and src/peft/tuners/lora/loraga.py line 101 call torch.load on config-specified cache/covariance files without weightsonly=True, bypassing peft's own safe-loading wrapper used elsewhere i...

EUVD-2026-53366

Hugging Face peft's LoRA-GA and CorDA initialization modules src/peft/tuners/lora/corda.py lines 102 and 163, and src/peft/tuners/lora/loraga.py line 101 call torch.load on config-specified cache/covariance files without weightsonly=True, bypassing peft's own safe-loading wrapper used elsewhere i...

CVE-2026-71281 peft Unsafe Deserialization via torch.load() Without weights_only in LoRA-GA and CorDA Modules

Hugging Face peft's LoRA-GA and CorDA initialization modules src/peft/tuners/lora/corda.py lines 102 and 163, and src/peft/tuners/lora/loraga.py line 101 call torch.load on config-specified cache/covariance files without weightsonly=True, bypassing peft's own safe-loading wrapper used elsewhere i...

CVE-2026-12484

A vulnerability in keras-team/keras version 3.15.0 allows unsafe deserialization of attacker-controlled PyTorch pickle data through the public keras.layers.TorchModuleWrapper.fromconfig method. This method invokes torch.load..., weightsonly=False without requiring an explicit unsafe opt-in, such ...

PyTorch Lightning allows arbitrary code execution through checkpoint _instantiator hyperparameters

PyTorch Lightning through 2.6.5, fixed in commit d710d68, contains a remote code execution vulnerability in the loadstate function that imports and executes attacker-controlled module names from checkpoint instantiator hyperparameters. Attackers can craft malicious checkpoint files that bypass...

PYSEC-2026-3624

PyTorch Lightning through 2.6.5, fixed in commit d710d68, contains a remote code execution vulnerability in the loadstate function that imports and executes attacker-controlled module names from checkpoint instantiator hyperparameters. Attackers can craft malicious checkpoint files that bypass...

CVE-2026-58659

PyTorch Lightning through 2.6.5, fixed in commit d710d68, contains a remote code execution vulnerability in the loadstate function that imports and executes attacker-controlled module names from checkpoint instantiator hyperparameters. Attackers can craft malicious checkpoint files that bypass...

EUVD-2026-44751

PyTorch Lightning through 2.6.5, fixed in commit d710d68, contains a remote code execution vulnerability in the loadstate function that imports and executes attacker-controlled module names from checkpoint instantiator hyperparameters. Attackers can craft malicious checkpoint files that bypass...

PYSEC-2026-3068 Snorkel Trainer.load uses an unsafe torch.load

The snorkel library thru v0.10.0 contains an insecure deserialization vulnerability CWE-502 in the Trainer.load method of the Trainer class. The method loads model checkpoint files using torch.load without enabling the security-restrictive weightsonly=True parameter. This default behavior allows...

PyTorch Lightning load_from_checkpoint has an insecure checkpoint deserialization

PyTorch-Lightning versions 2.6.0 and earlier contain an insecure deserialization vulnerability CWE-502 in the checkpoint loading mechanism. The LightningModule.loadfromcheckpoint method, which is commonly used to load saved model states, internally calls torch.load without setting the...

PT-2026-59144

The flash-attention training framework thru commit e724e2588cbe754beb97cf7c011b5e7e34119e62 2025-13-04 contains an insecure deserialization vulnerability CWE-502 in its checkpoint loading mechanism. The load checkpoint function in checkpoint.py and the checkpoint loading code in eval.py use...

PYSEC-2026-1856 PyTorch Vulnerable to Remote Code Execution via Untrusted Checkpoint Files

Summary A vulnerability in PyTorch's weightsonly unpickler allows an attacker to craft a malicious checkpoint file .pth that, when loaded with torch.load..., weightsonly=True, can corrupt memory and potentially lead to arbitrary code execution. Vulnerability Details The weightsonly=True unpickler...

PYSEC-2026-1578 lmdeploy vulnerable to Arbitrary Code Execution via Insecure Deserialization in torch.load()

Summary An insecure deserialization vulnerability exists in lmdeploy where torch.load is called without the weightsonly=True parameter when loading model checkpoint files. This allows an attacker to execute arbitrary code on the victim's machine when they load a malicious .bin or .pt model file...

PYSEC-2026-1576 LLaMA-Factory allows Code Injection through improper vhead_file safeguards

Summary A critical remote code execution vulnerability was discovered during the Llama Factory training process. This vulnerability arises because the vheadfile is loaded without proper safeguards, allowing malicious attackers to execute arbitrary malicious code on the host system simply by passi...

PYSEC-2026-405 Ludwig framework is vulnerable to insecure deserialization in its model serving component

The Ludwig framework thru 0.10.4 is vulnerable to insecure deserialization CWE-502 in its model serving component. When starting a model server with the ludwig serve command, the framework loads model weight files using torch.load without enabling the security-restrictive weightsonly=True...

CVE-2026-31229

A flaw was found in the Adversarial Robustness Toolbox ART, specifically within its Kubeflow component. This vulnerability, categorized as insecure deserialization CWE-502, occurs when loading model weights without proper security restrictions. A remote attacker can exploit this by uploading a...

CVE-2026-31224

The snorkel library thru v0.10.0 contains an insecure deserialization vulnerability CWE-502 in the MultitaskClassifier.load method of the MultitaskClassifier class. The method loads model weight files using torch.load without enabling the security-restrictive weightsonly=True parameter. This...