Skip to content

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.

Try to fix memory issue with lora.

comfyanonymous22f29d66cae4
diff --git a/comfy/model_management.py b/comfy/model_management.py
index b1afeb715..241706925 100644
--- a/comfy/model_management.py
+++ b/comfy/model_management.py
@@ -281,19 +281,23 @@ def load_model_gpu(model):
vram_set_state = VRAMState.LOW_VRAM
real_model = model.model
+ patch_model_to = None
if vram_set_state == VRAMState.DISABLED:
pass
elif vram_set_state == VRAMState.NORMAL_VRAM or vram_set_state == VRAMState.HIGH_VRAM or vram_set_state == VRAMState.SHARED:
model_accelerated = False
- real_model.to(torch_dev)
+ patch_model_to = torch_dev
try:
- real_model = model.patch_model()
+ real_model = model.patch_model(device_to=patch_model_to)
except Exception as e:
model.unpatch_model()
unload_model()
raise e
+ if patch_model_to is not None:
+ real_model.to(torch_dev)
+
if vram_set_state == VRAMState.NO_VRAM:
device_map = accelerate.infer_auto_device_map(real_model, max_memory={0: "256MiB", "cpu": "16GiB"})
accelerate.dispatch_model(real_model, device_map=device_map, main_device=torch_dev)
diff --git a/comfy/sd.py b/comfy/sd.py
index ddafa0b52..1f364dd1f 100644
--- a/comfy/sd.py
+++ b/comfy/sd.py
@@ -338,7 +338,7 @@ class ModelPatcher:
sd.pop(k)
return sd
- def patch_model(self):
+ def patch_model(self, device_to=None):
model_sd = self.model_state_dict()
for key in self.patches:
if key not in model_sd:
@@ -350,10 +350,13 @@ class ModelPatcher:
if key not in self.backup:
self.backup[key] = weight.to(self.offload_device)
- temp_weight = weight.to(torch.float32, copy=True)
+ if device_to is not None:
+ temp_weight = weight.float().to(device_to, copy=True)
+ else:
+ temp_weight = weight.to(torch.float32, copy=True)
out_weight = self.calculate_weight(self.patches[key], temp_weight, key).to(weight.dtype)
set_attr(self.model, key, out_weight)
- del weight
+ del temp_weight
return self.model
def calculate_weight(self, patches, weight, key):