PYSEC-2026-3938 vLLM: Incomplete CVE-2025-62164 remediation can be bypassed by concurrent prompt parts

Executive Summary The follow-up protection for CVE-2025-62164 is incomplete at vLLM revision 26587f9519e22a5c4549ead7595ad9ca3229c4fd. It wraps serialized prompt-embedding reconstruction and dense conversion in torch.sparse.checksparsetensorinvariants, but PyTorch 2.11.0 implements that context...

PT-2026-89723

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

CVE-2026-79785

X-AnyLabeling's model downloader disabled TLS certificate verification by calling ssl._create_unverified_context() in download_with_retry (within anylabeling/services/auto_labeling/model.py), so neither the certificate chain nor hostname was validated when fetching models over HTTPS. An attacker ...

EUVD-2026-65595

X-AnyLabeling's model downloader disabled TLS certificate verification. downloadwithretry in anylabeling/services/autolabeling/model.py built a context with ssl.createunverifiedcontext and passed it to urllib.request.urlopen, so neither the certificate chain nor the hostname was checked on any...

PT-2026-81392

X-AnyLabeling's model downloader disabled TLS certificate verification. download with retry in anylabeling/services/auto labeling/model.py built a context with ssl. create unverified context and passed it to urllib.request.urlopen, so neither the certificate chain nor the hostname was checked on...

SUSE CVE-2025-62164

vLLM is an inference and serving engine for large language models LLMs. From versions 0.10.2 to before 0.11.1, a memory corruption vulnerability could lead to a crash denial-of-service and potentially remote code execution RCE, exists in the Completions API endpoint. When processing user-supplied...

CVE-2026-73325

Fujitsu Research's OneCompression library before 1.2.1 contains an unsafe deserialization vulnerability that allows attackers to execute arbitrary code by supplying a crafted model.pt checkpoint file, as QuantizedModelLoader.loadquantizedmodelpt unconditionally calls torch.load with...

CVE-2026-73325

CVE-2026-73325 concerns Fujitsu Research’s OneCompression library 1.2.0, where an unsafe deserialization vulnerability arises when loading a crafted model.pt checkpoint. The root cause is QuantizedModelLoader.load_quantized_model_pt() unconditionally calling torch.load with weights_only=False, wh...

CVE-2026-73325

Fujitsu Research's OneCompression library 1.2.0 contains an unsafe deserialization vulnerability that allows attackers to execute arbitrary code by supplying a crafted model.pt checkpoint file, as QuantizedModelLoader.loadquantizedmodelpt unconditionally calls torch.load with weightsonly=False,...

CVE-2026-73325 Fujitsu OneCompression < 1.2.1 Arbitrary Code Execution via torch.load Deserialization

Fujitsu Research's OneCompression library before 1.2.1 contains an unsafe deserialization vulnerability that allows attackers to execute arbitrary code by supplying a crafted model.pt checkpoint file, as QuantizedModelLoader.loadquantizedmodelpt unconditionally calls torch.load with...

EUVD-2026-57313

Fujitsu Research's OneCompression library 1.2.0 contains an unsafe deserialization vulnerability that allows attackers to execute arbitrary code by supplying a crafted model.pt checkpoint file, as QuantizedModelLoader.loadquantizedmodelpt unconditionally calls torch.load with weightsonly=False,...

CVE-2026-73325 Fujitsu OneCompression < 1.2.1 Arbitrary Code Execution via torch.load Deserialization

Fujitsu Research's OneCompression library before 1.2.1 contains an unsafe deserialization vulnerability that allows attackers to execute arbitrary code by supplying a crafted model.pt checkpoint file, as QuantizedModelLoader.loadquantizedmodelpt unconditionally calls torch.load with...

CVE-2026-73325: Deserialization of Untrusted Data

Fujitsu Research's OneCompression library before 1.2.1 contains an unsafe deserialization vulnerability that allows attackers to execute arbitrary code by supplying a crafted model.pt checkpoint file, as QuantizedModelLoader.loadquantizedmodelpt unconditionally calls torch.load with...

PT-2026-71147

Fujitsu Research's OneCompression library 1.2.0 contains an unsafe deserialization vulnerability that allows attackers to execute arbitrary code by supplying a crafted model.pt checkpoint file, as QuantizedModelLoader.load quantized model pt unconditionally calls torch.load with weights only=Fals...

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: Deserialization of Untrusted Data

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: Deserialization of Untrusted Data

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...

CVE-2026-15531: Deserialization of Untrusted Data

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...

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...

CVE-2026-57516: Deserialization of Untrusted Data

Ray prior to 2.56.0 contains an unsafe deserialization vulnerability in the WebDataset reader that allows attackers to achieve remote code execution by supplying a malicious tar archive to the readwebdataset function. The defaultdecoder function in webdatasetdatasource.py unconditionally calls...