25 matches found
CVE-2026-15978
SGLang contains a model weight exfiltration vulnerability when no API keys are configured, as SGLang will expose two endpoints that allow a remote attacker to trigger distributed weight broadcasting using NCCL and then triggering data transfer, attackers can exfiltrate all model weights...
CVE-2026-15978 CVE-2026-15978
SGLang contains a model weight exfiltration vulnerability when no API keys are configured, as SGLang will expose two endpoints that allow a remote attacker to trigger distributed weight broadcasting using NCCL and then triggering data transfer, attackers can exfiltrate all model weights...
CVE-2026-15978 CVE-2026-15978
SGLang contains a model weight exfiltration vulnerability when no API keys are configured, as SGLang will expose two endpoints that allow a remote attacker to trigger distributed weight broadcasting using NCCL and then triggering data transfer, attackers can exfiltrate all model weights...
GHSA-3WW4-5JV9-J5GM vLLM's Artifact Pin Decay allows pinned deployments to load unpinned code, weights, and processors
Summary vLLM's revision pinning controls do not consistently apply to all artifacts loaded for a model. A deployment that supplies --revision or --code-revision can still load dynamic code, GGUF files, image processors, retrieval side weights, or same-repository subfolder weights/config from an...
model.weights.h5: h5py.ExternalLink at Group level silently followed during load_model(), bypassing CVE-2025-9905 fix — information disclosure from arbitrary HDF5 files
Keras 3.x introduced a fix for CVE-2025-9905 by checking dataset.external in H5IOStore.verifydataset. This check blocks datasets whose raw bytes are stored in external files via the HDF5 "External Data Storage" mechanism. However, HDF5 supports a second, unrelated external-reference mechanism:...
Vulnerabilities in Partial TEE-Shielded LLM Inference with Precomputed Noise
The deployment of large language models LLMs on third-party devices requires new ways to protect model intellectual property. While Trusted Execution Environments TEEs offer a promising solution, their performance limits can lead to a critical compromise: using a precomputed, static secret basis ...
CVE-2026-0897
A flaw was found in Keras. A remote attacker can cause a Denial of Service DoS by providing a specially crafted .keras archive containing a model weights file model.weights.h5 that declares an extremely large data shape. This can lead to excessive memory allocation, resulting in memory exhaustion...
UBUNTU-CVE-2026-0897
Allocation of Resources Without Limits or Throttling in the HDF5 weight loading component in Google Keras 3.0.0 through 3.13.0 on all platforms allows a remote attacker to cause a Denial of Service DoS through memory exhaustion and a crash of the Python interpreter via a crafted .keras archive...
CVE-2026-0897
Allocation of Resources Without Limits or Throttling in the HDF5 weight loading component in Google Keras 3.0.0 through 3.13.0 on all platforms allows a remote attacker to cause a Denial of Service DoS through memory exhaustion and a crash of the Python interpreter via a crafted .keras archive...
CVE-2026-0897 Denial of Service in Keras via Excessive Memory Allocation in HDF5 Metadata
Allocation of Resources Without Limits or Throttling in the HDF5 weight loading component in Google Keras 3.0.0 through 3.13.0 on all platforms allows a remote attacker to cause a Denial of Service DoS through memory exhaustion and a crash of the Python interpreter via a crafted .keras archive...
CVE-2026-0897
CVE-2026-0897 affects Google Keras (3.0.0–3.13.0) via the HDF5 weight loading component. A crafted .keras archive containing a valid model.weights.h5 file with an extremely large dataset shape can trigger memory exhaustion and crash the Python interpreter, causing a Denial of Service. Some connec...
CVE-2026-0897
Allocation of Resources Without Limits or Throttling in the HDF5 weight loading component in Google Keras 3.0.0 through 3.13.0 on all platforms allows a remote attacker to cause a Denial of Service DoS through memory exhaustion and a crash of the Python interpreter via a crafted .keras archive...
CVE-2026-0897 Denial of Service in Keras via Excessive Memory Allocation in HDF5 Metadata
Allocation of Resources Without Limits or Throttling in the HDF5 weight loading component in Google Keras 3.0.0 through 3.13.0 on all platforms allows a remote attacker to cause a Denial of Service DoS through memory exhaustion and a crash of the Python interpreter via a crafted .keras archive...
PT-2026-3014
Name of the Vulnerable Software and Affected Versions Keras versions 3.0.0 through 3.13.0 Description A flaw exists in the HDF5 weight loading component of Keras that allows for a denial of service. An attacker can provide a crafted .keras archive containing a valid model.weights.h5 file. This fi...
EUVD-2025-0192
Malicious code in bioql PyPI...
Dependacy chain attack through hijacking broken github repository at https://github.com/huggingface/transformers/blob/main/src/\ntransformers/models/fuyu/\nconvert_fuyu_model_weights_to_hf.py
Description Type: Dependency Chain Attack through hijacking broken github repository Risk: High Allows arbitrary code execution in model conversion workflows Affected Asset: https://github.com/adept-ai-labs/adept-inference Broken URL in Hugging Face Transformers Root Cause The Hugging Face...
ObfusBFA: a Holistic Approach to Safeguarding DNNs from Different Types of Bit-Flip Attacks
Bit-flip attacks BFAs represent a serious threat to Deep Neural Networks DNNs, where flipping a small number of bits in the model parameters or binary code can significantly degrade the model accuracy or mislead the model prediction in a desired way. Existing defenses exclusively focus on...
CVE-2022-29796
The HiAIserver has a vulnerability in verifying the validity of the weight used in the model.Successful exploitation of this vulnerability will affect AI services...