CVE-2026-1839Medium· 6.5▾ SunlitHuggingFace Transformers allows for arbitrary code execution in the `Trainer` class
▾ Sunlit zone — Low / medium · no exploitation signal
impact 35.8 · likelihood 0.1 · exploitation 0
Need a working PoC? Pro members can cast a request and our team develops one — it lands right here.
Exploit-prediction probability, daily snapshots since Jul 13.
Disclosure to exploitation, from the record and what we observed since indexing it.
Disclosed via OSV
0.3%
Last analysed / modified upstream
A vulnerability in the HuggingFace Transformers library, specifically in the Trainer class, allows for arbitrary code execution. The _load_rng_state() method in src/transformers/trainer.py at line 3059 calls torch.load() without the weights_only=True parameter. This issue affects all versions of the library supporting torch>=2.2 when used with PyTorch versions below 2.6, as the safe_globals() context manager provides no protection in these versions. An attacker can exploit this vulnerability by supplying a malicious checkpoint file, such as rng_state.pth, which can execute arbitrary code when loaded. The issue is resolved in version v5.0.0rc3.
transformers < 5.0.0rc3Upgrade to a patched release:
transformers 5.0.0rc3Connected by shared product, vendor, weakness, or advisory.
CVE-2024-11393High· 8.8Deserialization of Untrusted Data in Hugging Face Transformers
CVE-2024-11392High· 7.5Deserialization of Untrusted Data in Hugging Face Transformers
CVE-2024-11394High· 8.8Deserialization of Untrusted Data in Hugging Face Transformers
CVE-2026-9856High· 7.1A vulnerability in huggingface/transformers versions <=5.8.0.dev0 allows an attacker to perform arbitrary file writes via path traversal
CVE-2025-6051Medium· 5.3Hugging Face Transformers library has Regular Expression Denial of Service
CVE-2025-3263Medium· 5.3Transformers's ReDoS vulnerability in get_configuration_file can lead to catastrophic backtracking