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-2026-15976 CVE-2026-15976

SGLang contains a RCE vulnerability when attempting to load model weights from a HuggingFace repository, specifically within the /updateweightsfromdisk, where torch.load..., weightsonly=False fallback enables pickle deserialization of .bin files...

EUVD-2026-41089

Ray: Arbitrary code execution via ray.data.readwebdataset default decoder: pickle.loadsvalue and torch.loadweightsonly=False...

PYSEC-2026-3070 Snorkel MultitaskClassifier.load uses an unsafe torch.load

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...

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...

PYSEC-2026-2477 flash-attention contains an insecure deserialization vulnerability in its checkpoint loading mechanism

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

EUVD-2026-43277

A vulnerability has been found in yashbhalgat HashNeRF-pytorch up to 82885e698295982504eb6a26d060a6b2473e3706. Affected by this issue is the function torch.load of the file runnerf.py of the component Checkpoint File Handler. The manipulation of the argument ckptpath leads to deserialization. The...

CVE-2026-15531 yashbhalgat HashNeRF-pytorch Checkpoint File run_nerf.py torch.load deserialization

A vulnerability has been found in yashbhalgat HashNeRF-pytorch up to 82885e698295982504eb6a26d060a6b2473e3706. Affected by this issue is the function torch.load of the file runnerf.py of the component Checkpoint File Handler. The manipulation of the argument ckptpath leads to deserialization. The...

LLaMA-Factory Allows Arbitrary Code Execution via Unsafe Deserialization in Ilamafy_baichuan2.py

DescriptionA critical vulnerability exists in the llamafybaichuan2.py script of the LLaMA-Factory project. The script performs insecure deserialization using torch.load on user-supplied .bin files from an input directory. An attacker can exploit this behavior by crafting a malicious .bin file tha...

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...

EUVD-2025-210291

picklescan before 0.0.28 fails to detect malicious pickle files that invoke torch.utils.configmodule.loadconfig function within reduce methods. Attackers can craft pickle files embedding arbitrary code that evades detection but executes during pickle.load, enabling remote code execution in supply...

GHSA-CC5P-54X3-HCF8 Duplicate Advisory: Picklescan (scan_pytorch) Bypass via dynamic eval MAGIC_NUMBER

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-97f8-7cmv-76j2. This link is maintained to preserve external references. Original Description picklescan before 1.0.3 contains a scanning bypass vulnerability in the scanpytorch function that allows attackers to...

CVE-2026-53875

picklescan before 1.0.3 contains a scanning bypass vulnerability in the scanpytorch function that allows attackers to embed malicious magic numbers via dynamic eval using the reduce trick. Attackers can craft malicious PyTorch payloads that evade picklescan detection while remaining executable,...

CVE-2026-38950

An issue in ESA AnomalyMatch before 1.3.1 allow attackers to execute arbitrary code via crafted model checkpoint files. The affected components load model files from session directories using torch.load with unrestricted deserialization...

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...

CVE-2026-38950

An issue in ESA AnomalyMatch before 1.3.1 allow attackers to execute arbitrary code via crafted model checkpoint files. The affected components load model files from session directories using torch.load with unrestricted deserialization...

CVE-2026-38950

An issue in ESA AnomalyMatch before 1.3.1 allow attackers to execute arbitrary code via crafted model checkpoint files. The affected components load model files from session directories using torch.load with unrestricted deserialization...

CVE-2026-38950

CVE-2026-38950 affects ESA AnomalyMatch prior to 1.3.1. The issue arises from loading model files from session directories with torch.load(), enabling unrestricted deserialization and arbitrary code execution. Affected component: model checkpoint loading in AnomalyMatch. Impact: potential full co...

CVE-2026-38950

An issue in ESA AnomalyMatch before 1.3.1 allow attackers to execute arbitrary code via crafted model checkpoint files. The affected components load model files from session directories using torch.load with unrestricted deserialization...

CVE-2026-31222

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...