CVE-2025-14924
Deserialization RCE in Hugging Face Transformers megatron_gpt
Publication date: 2025-12-23
Last updated on: 2025-12-23
Assigner: Zero Day Initiative
Description
Description
CVSS Scores
EPSS Scores
| Probability: | |
| Percentile: |
Meta Information
Affected Vendors & Products
| Vendor | Product | Version / Range |
|---|---|---|
| hugging_face | transformers | 3.0 |
Helpful Resources
Exploitability
| CWE ID | Description |
|---|---|
| CWE-502 | The product deserializes untrusted data without sufficiently ensuring that the resulting data will be valid. |
Attack-Flow Graph
AI Powered Q&A
Can you explain this vulnerability to me?
This vulnerability is a remote code execution flaw in the Hugging Face Transformers library, specifically in the megatron_gpt2 model. It occurs because the software improperly validates data when parsing checkpoints, allowing deserialization of untrusted data. An attacker can exploit this by tricking a user into visiting a malicious webpage or opening a malicious file, which then lets the attacker execute arbitrary code within the affected process. [1]
How can this vulnerability impact me? :
If exploited, this vulnerability allows an attacker to execute arbitrary code on your system with the privileges of the affected process. This can lead to full compromise of confidentiality, integrity, and availability of your data and system, potentially allowing data theft, data manipulation, or service disruption. [1]
What immediate steps should I take to mitigate this vulnerability?
To mitigate this vulnerability, avoid opening untrusted files or visiting untrusted webpages that may contain malicious checkpoints. Ensure that any checkpoint files loaded into the Hugging Face Transformers megatron_gpt2 model are from trusted sources. Applying any available patches or updates from the Hugging Face Transformers library that address this deserialization issue is recommended once released. [1]