CVE-2025-62609

HighCVSS 7.5

Published: 21/11/2025(UTC)Updated: 08/10/2026(UTC)Translated: 08/10/2026(UTC)NVD NIST

Exploitation Probability (EPSS)

Low risk

0.38%

30th percentile - higher than 30% of all known CVEs

Summary

MLX is an array framework for machine learning on Apple silicon. Prior to version 0.29.4, there is a segmentation fault in mlx::core::load_gguf() when loading malicious GGUF files. An untrusted pointer from the external gguflib library is dereferenced without validation, causing application crash. This issue has been patched in version 0.29.4.

Risk Assessment

A malicious GGUF file can crash an application using MLX, leading to denial of service. The risk applies to environments processing untrusted model files.

Recommendation

Update MLX to version 0.29.4 or later. Avoid loading GGUF files from untrusted sources.

Other vulnerabilities in MLX

CVE-2025-62608Critical
MLX, an array framework for machine learning on Apple silicon, prior to version 0.29.4 has a heap buffer overflow in mlx::core::load() when parsing malicious NumPy .npy files. An attacker-controlled file causes a 13-byte out-of-bounds read, leading to crash or information disclosure.

Original NVD description (English source)

MLX is an array framework for machine learning on Apple silicon. Prior to version 0.29.4, there is a segmentation fault in mlx::core::load_gguf() when loading malicious GGUF files. Untrusted pointer from external gguflib library is dereferenced without validation, causing application crash. This issue has been patched in version 0.29.4.

Vulnerability data from NVD (NIST) · CISA KEV · EPSS