CriticalVulnerability
CVE-2025-50472: modelscope/ms-swift arbitrary code execution via deserialization
- Source
- NVD(opens in a new tab)
- Identifier
- CVE-2025-50472
- Published
- Record updated
- Known exploitation
- Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
- EPSS
- 1.5%
Summary
The modelscope/ms-swift library through 2.6.1 is vulnerable to arbitrary code execution. The `load_model_meta()` function of the `ModelFileSystemCache()` class calls `pickle.load()` on untrusted data, so a crafted serialized `.mdl` payload runs code when a victim loads what appears to be a normal checkpoint during training. The payload file is hidden and the training run continues normally, so the victim gets no visible sign of the execution.
Mitigation
The source does not state a fix yet. Check the original advisory for updates.
Related items
- HighCVE-2026-119120: Crawl4AI Docker API lets clients choose the HuggingFace model it loadsSame vendor · NVD/CVE Database
- HighGHSA-wf6j-7hvr-gr93: Hydra legacy target blacklist permits unsafe PyYAML loadingSame vendor · Hugging Face Security Advisories
- MediumCVE-2026-100653: vLLM unpinned Hugging Face artifact loads for FunAudioChat and Tarsier2Same vendor · NVD/CVE Database
- HighGHSA-3hmm-rh5q-gwwr: LMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loadingSame vendor · Hugging Face Security Advisories
- MediumCVE-2026-80047: Hugging Face Transformers writes remote Python files to disk before trust checkSame vendor · NVD/CVE Database