Ryan Dick da589b3f1f Memory optimization to load state dicts one module at a time in CachedModelWithPartialLoad when we are not storing a CPU copy of the state dict (i.e. when keep_ram_copy_of_weights=False).

2025-01-16 17:00:33 +00:00

..

model_cache

Memory optimization to load state dicts one module at a time in CachedModelWithPartialLoad when we are not storing a CPU copy of the state dict (i.e. when keep_ram_copy_of_weights=False).

2025-01-16 17:00:33 +00:00

model_loaders

Update the model loading logic for several of the large FLUX-related models to ensure that the model is initialized on the meta device prior to loading the state dict into it. This helps to keep peak memory down.

2025-01-16 02:30:28 +00:00

__init__.py

Rename model_cache_default.py -> model_cache.py.

2024-12-24 14:23:18 +00:00

load_base.py

Allow expensive operations to request more working memory.

2025-01-07 01:20:13 +00:00

load_default.py

Get rid of ModelLocker. It was an unnecessary layer of indirection.

2024-12-24 14:23:18 +00:00

memory_snapshot.py

Add ruff rule to disallow relative parent imports.

2024-07-04 09:35:37 -04:00

model_loader_registry.py

Apply ruff rule to disallow all relative imports.

2024-07-04 09:35:37 -04:00

model_util.py

Rename LoRAModelRaw to ModelPatchRaw.

2024-12-17 13:20:19 +00:00

optimizations.py

final tidying before marking PR as ready for review

2024-03-01 10:42:33 +11:00