Description
So i have a 3090 and 16gb of system ram, im getting 64gb tomorrow.
Big issue im facing atm is NO MATTER THE MODEL I TRY to load from the 14b version i always OOM especially when i try to load a Lora i oom... i tired block swapping and that also fails.
`During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "threading.py", line 1075, in _bootstrap_inner
File "threading.py", line 1012, in run
File "G:\AI Stuff\ComfyUI_windows_portable\ComfyUI\main.py", line 175, in prompt_worker
e.execute(item[2], prompt_id, item[3], item[4])
File "G:\AI Stuff\ComfyUI_windows_portable\ComfyUI\execution.py", line 542, in execute
result, error, ex = execute(self.server, dynamic_prompt, self.caches, node_id, extra_data, executed, prompt_id, execution_list, pending_subgraph_results)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "G:\AI Stuff\ComfyUI_windows_portable\ComfyUI\execution.py", line 428, in execute
input_data_formatted[name] = [format_value(x) for x in inputs]
^^^^^^^^^^^^^^^
File "G:\AI Stuff\ComfyUI_windows_portable\ComfyUI\execution.py", line 280, in format_value
return str(x)
^^^^^^
File "G:\AI Stuff\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch_tensor.py", line 590, in repr
return torch._tensor_str._str(self, tensor_contents=tensor_contents)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "G:\AI Stuff\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch_tensor_str.py", line 710, in _str
return _str_intern(self, tensor_contents=tensor_contents)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "G:\AI Stuff\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch_tensor_str.py", line 631, in _str_intern
tensor_str = _tensor_str(self, indent)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "G:\AI Stuff\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch_tensor_str.py", line 363, in _tensor_str
formatter = _Formatter(get_summarized_data(self) if summarize else self)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "G:\AI Stuff\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch_tensor_str.py", line 399, in get_summarized_data
return torch.stack([get_summarized_data(x) for x in (start + end)])
^^^^^^^^^^^^^^^^^^^^^^
File "G:\AI Stuff\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch_tensor_str.py", line 401, in get_summarized_data
return torch.stack([get_summarized_data(x) for x in self])
^^^^^^^^^^^^^^^^^^^^^^
File "G:\AI Stuff\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch_tensor_str.py", line 399, in get_summarized_data
return torch.stack([get_summarized_data(x) for x in (start + end)])
^^^^^^^^^^^^^^^^^^^^^^
File "G:\AI Stuff\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch_tensor_str.py", line 389, in get_summarized_data
return torch.cat(
^^^^^^^^^^
RuntimeError: CUDA error: out of memory
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.`
i was using a gguf model 14b and the self forcing lora rendering a 480p i2v will take me like 40 seconds on 3 samples which actually gives me a pretty good decent result...
on wrapper i cant even load any model. help pls, workflow in image