CVE-2025-5197

published 6 Aug 2025 · updated 6 Aug 2025

Regular Expression Denial of Service (ReDoS) in huggingface/transformers

5.3MediumCVSS 5.3EPSS 0%

No exploitation in the wild has been reported. EPSS puts the chance of exploitation in the next 30 days at 0%. Fixed in 4.53.0.

A Regular Expression Denial of Service (ReDoS) vulnerability exists in the Hugging Face Transformers library, specifically in the `convert_tf_weight_name_to_pt_weight_name()` function. This function, responsible for converting TensorFlow weight names to PyTorch format, uses a regex pattern `/[^/]*___([^/]*)/` that can be exploited to cause excessive CPU consumption through crafted input strings due to catastrophic backtracking. The vulnerability affects versions up to 4.51.3 and is fixed in version 4.53.0. This issue can lead to service disruption, resource exhaustion, and potential API service vulnerabilities, impacting model conversion processes between TensorFlow and PyTorch formats.

Affected

Vendors, products and packages with version ranges

VendorProductVersionsFixed
huggingface<4.53.04.53.0

Timeline

Every event on this vulnerability

Embed

A live severity badge for an advisory, README or ticket

CVE-2025-5197 severity on Zeptric[![CVE-2025-5197](https://zeptric.io/badge/cve/CVE-2025-5197.svg)](https://zeptric.io/cve/CVE-2025-5197)