CVE Feed

    Dashboard / CVE / CVE-2025-23306

    CVE-2025-23306

    NVIDIA Megatron-LM for all platforms contains a vulnerability in the megatron/training/ arguments.py component where an attacker could cause a code injection issue by providing a malicious input. A successful exploit of this vulnerability may lead to code execution, escalation of privileges, information disclosure, and data tampering.

    Published:Aug 13, 2025
    Last Modified:Sep 19, 2025
    EPS:Aug 13, 2025
    EPSS Score:0.00024
    CVSS Score:7.8

    Affected Products

    Vendor
    Nvidia
    Product
    Megatron-lm

    Exploits

    No exploit reference

    Related CVEs

    CVE-2026-24152 (CVSS 7.8) — NVIDIA Megatron-LM contains a vulnerability in checkpoint loading where an Attacker may cause an RCE by convin…
    CVE-2026-24151 (CVSS 7.8) — NVIDIA Megatron-LM contains a vulnerability in inferencing where an Attacker may cause an RCE by convincing a …
    CVE-2026-24150 (CVSS 7.8) — NVIDIA Megatron-LM contains a vulnerability in checkpoint loading where an Attacker may cause an RCE by convin…
    CVE-2025-33248 (CVSS 7.8) — NVIDIA Megatron-LM contains a vulnerability in the hybrid conversion script where an Attacker may cause an RCE…
    CVE-2025-33247 (CVSS 7.8) — NVIDIA Megatron LM contains a vulnerability in quantization configuration loading, which could allow remote co…
    CVE-2026-24149 (CVSS 7.8) — NVIDIA Megatron-LM for all platforms contains a vulnerability in a script, where malicious data created by an …
    CVE-2025-23357 (CVSS 7.8) — NVIDIA Megatron-LM for all platforms contains a vulnerability in a script, where malicious data created by an …
    CVE-2025-23354 (CVSS 7.8) — NVIDIA Megatron-LM for all platforms contains a vulnerability in the ensemble_classifer script where malicious…
    CVE-2025-23353 (CVSS 7.8) — NVIDIA Megatron-LM for all platforms contains a vulnerability in the msdp preprocessing script where malicious…
    CVE-2025-23349 (CVSS 7.8) — NVIDIA Megatron-LM for all platforms contains a vulnerability in the tasks/orqa/unsupervised/nq.py component, …

    Common Vulnerability Scoring System

    Attack Vector
    Network
    Adjacent
    Local
    Physical
    Privileges Required
    None
    Low
    High
    User Interaction
    None
    Required
    Scope
    Unchanged
    Changed
    Confidentiality
    None
    Low
    High
    Integrity
    None
    Low
    High
    Availability
    None
    Low
    High