CVE-2026-31253: flash-attention contains an insecure deserialization vulnerability in its checkpoint loading mechanism
The flash-attention training framework thru commit e724e2588cbe754beb97cf7c011b5e7e34119e62 (2025-13-04) contains an insecure deserialization vulnerability (CWE-502) in its checkpoint loading mechanism. The loadcheckpoint() function in checkpoint.py and the checkpoint loading code in eval.py use torch.load() without enabling the security-restrictive weightsonly=True parameter.
Published May 11, 2026 · Updated Jun 30, 2026
Affected packages
flash_attn
- Ecosystem
- PyPI
- Affected
- <= 2.8.3
- Fixed in
- No patched version yet
- Check
pip show flash_attn
Advisory details
From the GitHub Advisory Database (CC BY 4.0).
The flash-attention training framework thru commit e724e2588cbe754beb97cf7c011b5e7e34119e62 (2025-13-04) contains an insecure deserialization vulnerability (CWE-502) in its checkpoint loading mechanism. The load_checkpoint() function in checkpoint.py and the checkpoint loading code in eval.py use torch.load() without enabling the security-restrictive weights_only=True parameter. This allows the deserialization of arbitrary Python objects via the pickle module. An attacker can exploit this by providing a maliciously crafted checkpoint file. When a victim loads this checkpoint during model warmstarting or evaluation, arbitrary code is executed on the victim's system.
Credited to warsang.
References
How this page is made: the data panel comes from the GitHub Advisory Database, CISA’s KEV catalog, the NVD and FIRST EPSS, and updates twice a day. No analysis has been written for this entry yet. Methodology
