LLama Factory enables fine-tuning of large language models. Prior to version 1.0.0, a critical vulnerability exists in the `llamafy_baichuan2.py` script of the LLaMA-Factory project. The script performs insecure deserialization using `torch.load()` on user-supplied `.bin` files from an input directory. An attacker can exploit this behavior by crafting a malicious `.bin` file that executes arbitrary commands during deserialization. This issue has been patched in version 1.0.0.
2025-05-01T18:15:58.117
2025-06-17T14:19:39.290
Analyzed
CVSSv3.1: 6.1 (MEDIUM)
| Type | Vendor | Product | Version/Range | Vulnerable? |
|---|---|---|---|---|
| Application | hiyouga | llama-factory | < 1.0.0 | Yes |