CVE-2026-76841

Xinference (by XorbitsAI) through version 2.11.0 unconditionally sets trust_remote_code=True at six model loader call sites spanning the rerank, embedding, and LLM modules. An attacker with model launch access can register a model with an unknown type and an arbitrary path, causing AutoTokenizer....

PT-2026-80851

Xinference loads models with Hugging Face remote code execution unconditionally enabled, and before version 2.12.0 exposes no setting to disable it. Six loader call sites pass trust remote code=True as a literal or as an unconditional default: RerankModel. get tokenizer in...

CVE-2026-76395 Remote Code Execution (RCE) through Deserialization of Untrusted Data in the Model Loading REST API in Splunk AI Toolkit

In Splunk AI Toolkit versions below 6.0.0, a user who holds the "power" Splunk role could execute arbitrary code on the Splunk server by loading a model file containing crafted sparse matrix data. The deserialization of untrusted data is possible because a model codec in Splunk AI Toolkit...

CVE-2026-76395

Splunk AI Toolkit versions below 6.0.0 deserialize sparse matrix data containing embedded pickle payloads without validation, allowing a user with the "power" role to execute arbitrary code via the model-loading REST API. The fix is to upgrade to version 6.0.0 or later. No in-the-wild exploitatio...

CVE-2026-76395 Remote Code Execution (RCE) through Deserialization of Untrusted Data in the Model Loading REST API in Splunk AI Toolkit

In Splunk AI Toolkit versions below 6.0.0, a user who holds the "power" Splunk role could execute arbitrary code on the Splunk server by loading a model file containing crafted sparse matrix data. The deserialization of untrusted data is possible because a model codec in Splunk AI Toolkit...

CVE-2026-73067

A flaw was found in Tesseract, an open-source Optical Character Recognition OCR engine. A remote attacker could exploit this vulnerability by providing a specially crafted .traineddata model. When this malicious model is loaded, it can cause the software to read beyond its allocated memory, leadi...

Keras model loading is vulnerable to denial of service through HDF5 shape bombs

A vulnerability in keras-team/keras versions 3.15.0 allows for a denial of service DoS attack when loading malicious .keras model files via the keras.models.loadmodel function. The H5IOStore.getitem method in keras/src/saving/savinglib.py does not validate the shape or size of datasets, leading t...

GHSA-74M6-M3XX-3VMJ Keras model loading is vulnerable to denial of service through HDF5 shape bombs

A vulnerability in keras-team/keras versions 3.15.0 allows for a denial of service DoS attack when loading malicious .keras model files via the keras.models.loadmodel function. The H5IOStore.getitem method in keras/src/saving/savinglib.py does not validate the shape or size of datasets, leading t...

PT-2026-69334

Name of the Vulnerable Software and Affected Versions keras-team/keras versions prior to 3.15.1 Description A denial of service DoS issue occurs when loading malicious .keras model files using the keras.models.load model function. The H5IOStore. getitem method in keras/src/saving/saving lib.py...

Linux Distros Unpatched Vulnerability : CVE-2026-12570

The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - A vulnerability in keras-team/keras versions = 3.15.0 allows for a denial of service DoS attack when loading malicious .keras model files via the...

CVE-2026-68770

sentence-transformers contains a security control bypass vulnerability that allows attackers to achieve arbitrary code execution by exploiting a logic flaw in the importmoduleclass helper within sentencetransformers/util/misc.py, where the guard condition includes an 'or...

CVE-2026-65920

A flaw was found in Diffusers. This path traversal vulnerability in the getcheckpointshardfiles function allows a remote attacker to read arbitrary files. By supplying malicious weightmap values in model index JSON, an attacker can use ../ sequences or absolute paths to escape the model directory...

CVE-2026-65920 Diffusers Path Traversal via weight_map Arbitrary File Read

Diffusers through 0.39.0, fixed in commit cee298c, contains a path traversal vulnerability in the getcheckpointshardfiles function that allows attackers to read arbitrary files by supplying malicious weightmap values in model index JSON. Attackers can use ../ sequences or absolute paths in...

PYSEC-2026-3498 PocketSphinx: Buffer overflows in language and acoustic model loading code

Impact The trie language model code introduced in PocketSphinx 5prealpha failed to check various boundary conditions when reading the headers of ARPA, DMP, and binary format language model files. In the case of invalid, corrupted or malicious input files, this could lead to stack and heap buffer...

PT-2026-64106

Impact The trie language model code introduced in PocketSphinx 5prealpha failed to check various boundary conditions when reading the headers of ARPA, DMP, and binary format language model files. In the case of invalid, corrupted or malicious input files, this could lead to stack and heap buffer...

Heap-based Buffer Overflow

Overview pocketsphinx is an Official Python bindings for PocketSphinx Affected versions of this package are vulnerable to Heap-based Buffer Overflow via the model loading process. An attacker can execute arbitrary code or cause a crash by supplying a crafted or malicious language or acoustic mode...

PT-2026-60899

Name of the Vulnerable Software and Affected Versions PocketSphinx versions 5prealpha through 5.1.0 Description The trie language model code fails to check boundary conditions when reading headers of ARPA, DMP, and binary format language model files, which can lead to stack and heap buffer...

CVE-2026-50144 ncnn: Out-of-bounds heap write in ParamDict::load_param via unchecked negative parameter id

ncnn is a high-performance neural network inference framework optimized for the mobile platform. In commit e54f7b1f88434e1d844ea0551b880a1cfb079ce1 and earlier, ncnn allows an out-of-bounds heap write in ncnn::ParamDict::loadparam when Net::loadparam loads a malicious .param model file because th...

OpenMed vulnerable to remote code injection through privacy-filter model loading path

OpenMed before 1.5.2 contains a remote code execution vulnerability in the PII privacy-filter model loading path. The privacy-filter dispatcher used broad substring matching on the user-supplied modelname parameter, allowing a value such as attacker/foo-privacy-filter-bar to route through a path...

PYSEC-2026-2852 OpenMed vulnerable to remote code injection through privacy-filter model loading path

OpenMed before 1.5.2 contains a remote code execution vulnerability in the PII privacy-filter model loading path. The privacy-filter dispatcher used broad substring matching on the user-supplied modelname parameter, allowing a value such as attacker/foo-privacy-filter-bar to route through a path...