CVE-2025-62609: MLX segmentation fault in GGUF file loading from malicious files

Summary

CVE-2025-62609 affects MLX, an array framework for machine learning on Apple silicon, before version 0.29.4. Loading a malicious GGUF file through mlx::core::load_gguf() causes a segmentation fault, because an untrusted pointer from the external gguflib library is dereferenced without validation, crashing the application. The CNA, GitHub, Inc., rates it 5.5 MEDIUM under CVSS 4.0 with a network attack vector, no privileges required, and a low availability impact.

Mitigation

Fixed in version 0.29.4.

Affected packages in the Exposure Registry

Matched by package name and ecosystem. Each entry shows whether the package delegates to a language model and how many tracked packages depend on it.

• mlxPyPINo LLM dependency detected · 8 tracked dependents

Related items

• MediumCVE-2026-121936: easyVoice server-side request forgery through openaiBaseUrl in /api/v1/ttsSimilar attack · NVD/CVE Database
• MediumCVE-2026-118628: MLflow get-artifacts server-side request forgery information disclosureSimilar attack · NVD/CVE Database
• HighCVE-2026-118625: MLflow Tracking Server privilege escalation via get_endpointsSimilar attack · NVD/CVE Database
• InfoWhen trust becomes the payload: Inside the threats, etc quarterly reportSimilar attack · Elastic Security Labs
• MediumCritical Langflow flaws enable authentication bypass and RCESimilar attack · CSO Online