The _load_model() function in the neural_magic_training.py script of the optimate project in commit a6d302f912b481c94370811af6b11402f51d377f (2024-07-21) allows arbitrary code execution. When a user supplies a directory path via the --model command-line argument, the function reads a module.py file from that directory and executes its contents directly using Python's exec() function. This design does not validate or sanitize the file's content, allowing an attacker who controls the input directory to execute arbitrary Python code in the context of the process running the script.
Fix
Patch, vendor advisory and mitigation references on NVD 1
- https://www.notion.so/CVE-2026-31217-35d1e13931888179ae40dea5258d2db9 Mitigation, Third Party Advisory
Every line above is quoted from NVD, CISA, Microsoft, the vendor's own advisory or DISA; nothing is generated.
Change history 1
- First seenCVSS 9.8 · nebulyNVD modified 2026-10-19 22:10
Affected (CPE criteria) 1
cpe:2.3:a:nebuly:optimate:2024-07-21:*:*:*:*:*:*:*
References 2
- https://github.com/nebuly-ai/optimate Product
- https://www.notion.so/CVE-2026-31217-35d1e13931888179ae40dea5258d2db9 Mitigation, Third Party Advisory