Artiprocher
commented
3 hours ago You need more GPU VRAM. Or reduce the resolution.
Open Hing-Hing-Zone-Zone opened 2 days ago
Artiprocher
commented
3 hours ago You need more GPU VRAM. Or reduce the resolution.
Hello ~ I got this error when doing Video Creator... How can I solve this out?
Loading videos ...
Loading videos ... done!
Loading models ...
Loading models ... done!
Synthesizing videos ...
OutOfMemoryError: CUDA out of memory. Tried to allocate 480.00 MiB. GPU Traceback: File "C:\Users\user\AppData\Local\Programs\Python\Python310\lib\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 589, in _run_script exec(code, module.dict) File "C:\AI\DiffSynth-Studio\pages\2_Video_Creator.py", line 197, in
SDVideoPipelineRunner(in_streamlit=True).run(config)
File "C:\AI\DiffSynth-Studio\diffsynth\pipelines\stable_diffusion_video.py", line 349, in run
output_video = self.synthesize_video(model_manager, pipe, config["pipeline"]["seed"], smoother, config["pipeline"]["pipeline_inputs"])
File "C:\AI\DiffSynth-Studio\diffsynth\pipelines\stable_diffusion_video.py", line 296, in synthesize_video
output_video = pipe(pipeline_inputs, smoother=smoother, progress_bar_st=progress_bar_st)
File "C:\Users\user\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, kwargs)
File "C:\AI\DiffSynth-Studio\diffsynth\pipelines\stable_diffusion_video.py", line 221, in call
noise_pred_posi = lets_dance_with_long_video(
File "C:\AI\DiffSynth-Studio\diffsynth\pipelines\stable_diffusion_video.py", line 38, in lets_dance_with_long_video
hidden_states_batch = lets_dance(
File "C:\AI\DiffSynth-Studio\diffsynth\pipelines\dancer.py", line 93, in lets_dance
hidden_states, time_emb, text_emb, res_stack = motion_modules.motion_modules[motion_module_id](
File "C:\Users\user\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, *kwargs)
File "C:\Users\user\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl
return forward_call(args, kwargs)
File "C:\AI\DiffSynth-Studio\diffsynth\models\sd_motion.py", line 82, in forward
hidden_states = block(
File "C:\Users\user\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, kwargs)
File "C:\Users\user\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl
return forward_call(*args, *kwargs)
File "C:\AI\DiffSynth-Studio\diffsynth\models\sd_motion.py", line 45, in forward
ff_output = self.act_fn(norm_hidden_states)
File "C:\Users\user\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl
return self._call_impl(args, kwargs)
File "C:\Users\user\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl
return forward_call(*args, kwargs)
File "C:\AI\DiffSynth-Studio\diffsynth\models\sd_unet.py", line 27, in forward
hidden_states, gate = self.proj(hidden_states).chunk(2, dim=-1)
File "C:\Users\user\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, *kwargs)
File "C:\Users\user\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl
return forward_call(args, kwargs)
File "C:\Users\user\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\nn\modules\linear.py", line 116, in forward
return F.linear(input, self.weight, self.bias)