ONNX is an open-source machine learning model interchange format that enables portability of trained models across different frameworks and platforms. The vendor's disclosed vulnerabilities center on its core runtime implementation and cluster around path-traversal weaknesses that can arise during model serialization and file handling operations. Current exploitation activity and exposure counts are shown alongside this summary.
The number and severity of CVEs published that impact products developed by Onnx over time
Signals from CVEs in this vendor scope (1 CVEs).
1 CVEs · Highest risk first
| CVE | Published | CVSS | Risk | KEV | Exploit |
|---|---|---|---|---|---|
CVE-2024-7776CRITICAL A vulnerability in the `download_model` function of the onnx/onnx framework, before and including version 1.16.1, allows for arbitrary file overwrite due to inadequate prevention o | Mar 20, 2025 | 9.1 | 25 | NO | NO |
Signals from CVEs in this vendor scope (1 CVEs).
An overview of all social media posts that mention a CVE ID that affects a product developed by Onnx.
Media articles that mention a CVE ID that affects a product developed by Onnx — matched by CVE ID, not by vendor name.