52 matches found
Deserialization of Untrusted Data
Overview keras is a Keras is a high-level neural networks API for Python.. Affected versions of this package are vulnerable to Deserialization of Untrusted Data via the function savinglib.loadmodel, which identifies the model source file, directory, or Hugging Face repository and then calls the...
CVE-2025-8747
CVE-2025-8747 corresponds to a safe-mode bypass in Keras Model.load_model, allowing arbitrary code execution by loading a crafted .keras archive. Connected IBM bulletins confirm the vulnerability affects Keras 3.0.0–3.10.0 and describe a bypass via manipulated config.json or inner Lambda mechanis...
CVE-2025-8747 Keras safe_mode bypass allows arbitrary code execution when loading a malicious model.
A safe mode bypass vulnerability in the Model.loadmodel method in Keras versions 3.0.0 through 3.10.0 allows an attacker to achieve arbitrary code execution by convincing a user to load a specially crafted .keras model archive...
Deserialization of Untrusted Data
Overview ms-swift is a Swift: Scalable lightWeight Infrastructure for Fine-Tuning Affected versions of this package are vulnerable to Deserialization of Untrusted Data via the loadmodelmeta function. An attacker can execute arbitrary code by supplying a maliciously crafted serialized .mdl file th...
SWIFT 安全漏洞
SWIFT is a large model and multimodal large model fine-tuning deployment framework from ModelScope open source. A security vulnerability exists in SWIFT 2.6.1 and earlier versions, which stems from the deserialization of untrustworthy data by the loadmodelmeta function in the ModelFileSystemCache...
The vulnerability of the Model.load_model function in the Keras library allows a hacker to execute arbitrary code.
The vulnerability of the Model.loadmodel function in the Keras library is related to incorrect code generation. Exploiting this vulnerability could allow an attacker to execute arbitrary code...
Deserialization of Untrusted Data
Overview keras is a Keras is a high-level neural networks API for Python.. Affected versions of this package are vulnerable to Deserialization of Untrusted Data in the loadmodel function, which uses pickle for serialization. An attacker can execute code even if safemode is set to True, by supplyi...
OSV-2024-1090 UNKNOWN READ in ggml_free
OSS-Fuzz report: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=42538517 Crash type: UNKNOWN READ Crash state: ggmlfree llamamodel::llamamodel llamaloadmodelfromfile...
PT-2024-40576
Name of the Vulnerable Software and Affected Versions: Llama affected versions not specified Description: The issue is related to a crash caused by an unknown read. Technical details about the crash include the ggml free function, the llama model destructor, and the llama load model from file...
PT-2024-18273 · Van Der Schaar · Van Der Schaar Lab Autoprognosis
Name of the Vulnerable Software and Affected Versions: van der Schaar LAB AutoPrognosis version 0.1.21 Description: A critical vulnerability was found in the function load model from file of the component Release Note Handler. The manipulation leads to deserialization. The attack can be initiated...
SUSE CVE-2020-14212
FFmpeg through 4.3 has a heap-based buffer overflow in aviogetstr in libavformat/aviobuf.c because dnnbackendnative.c calls ffdnnloadmodelnative and a certain index check is omitted...