Claude f71b604438 security: add weights_only=True to all torch.load() calls ... Fixes a critical security vulnerability where malicious model checkpoints could execute arbitrary code through pickle deserialization. Changes: - wan/modules/vae.py: Add weights_only=True to torch.load() - wan/modules/clip.py: Add weights_only=True to torch.load() - wan/modules/t5.py: Add weights_only=True to torch.load() This prevents arbitrary code execution when loading untrusted checkpoints while maintaining full compatibility with legitimate model weights. Security Impact: Critical - prevents RCE attacks Breaking Changes: None - weights_only=True is compatible with all standard PyTorch state_dict files		2025-11-19 04:24:14 +00:00
..
__init__.py	[feature] Add VACE (#389)	2025-05-14 20:44:25 +08:00
attention.py	init upload	2025-02-25 22:07:47 +08:00
clip.py	security: add weights_only=True to all torch.load() calls	2025-11-19 04:24:14 +00:00
model.py	Format the code (#402)	2025-05-16 12:35:38 +08:00
t5.py	security: add weights_only=True to all torch.load() calls	2025-11-19 04:24:14 +00:00
tokenizers.py	init upload	2025-02-25 22:07:47 +08:00
vace_model.py	Format the code (#402)	2025-05-16 12:35:38 +08:00
vae.py	security: add weights_only=True to all torch.load() calls	2025-11-19 04:24:14 +00:00
xlm_roberta.py	init upload	2025-02-25 22:07:47 +08:00