search

rhymes-ai / Allegro

Allegro is a powerful text-to-video model that generates high-quality videos up to 6 seconds at 15 FPS and 720p resolution from simple text input.
https://rhymes.ai/
Apache License 2.0
291 stars 14 forks source link

RuntimeError #19

Closed CHK-0000 closed 19 hours ago

CHK-0000 commented 20 hours ago

Traceback (most recent call last): File "C:\Allegro\single_inference.py", line 99, in single_inference(args) File "C:\Allegro\single_inference.py", line 65, in single_inference out_video = allegro_pipeline( File "C:\Users\ksl10\anaconda3\envs\test\lib\site-packages\torch\utils_contextlib.py", line 116, in decorate_context return func(*args, kwargs) File "C:\Allegro\allegro\pipelines\pipeline_allegro.py", line 773, in call noise_pred = self.transformer( File "C:\Users\ksl10\anaconda3\envs\test\lib\site-packages\torch\nn\modules\module.py", line 1553, in _wrapped_call_impl return self._call_impl(*args, *kwargs) File "C:\Users\ksl10\anaconda3\envs\test\lib\site-packages\torch\nn\modules\module.py", line 1562, in _call_impl return forward_call(args, kwargs) File "C:\Users\ksl10\anaconda3\envs\test\lib\site-packages\accelerate\hooks.py", line 169, in new_forward output = module._old_forward(*args, kwargs) File "C:\Allegro\allegro\models\transformers\transformer_3d_allegro.py", line 331, in forward hidden_states = block( File "C:\Users\ksl10\anaconda3\envs\test\lib\site-packages\torch\nn\modules\module.py", line 1553, in _wrapped_call_impl return self._call_impl(*args, *kwargs) File "C:\Users\ksl10\anaconda3\envs\test\lib\site-packages\torch\nn\modules\module.py", line 1562, in _call_impl return forward_call(args, kwargs) File "C:\Users\ksl10\anaconda3\envs\test\lib\site-packages\accelerate\hooks.py", line 169, in new_forward output = module._old_forward(*args, kwargs) File "C:\Allegro\allegro\models\transformers\block.py", line 1093, in forward attn_output = self.attn1( File "C:\Users\ksl10\anaconda3\envs\test\lib\site-packages\torch\nn\modules\module.py", line 1553, in _wrapped_call_impl return self._call_impl(*args, *kwargs) File "C:\Users\ksl10\anaconda3\envs\test\lib\site-packages\torch\nn\modules\module.py", line 1562, in _call_impl return forward_call(args, kwargs) File "C:\Allegro\allegro\models\transformers\block.py", line 553, in forward return self.processor( File "C:\Allegro\allegro\models\transformers\block.py", line 824, in call hidden_states = F.scaled_dot_product_attention( RuntimeError: No available kernel. Aborting execution.

CUDA 11.8, TORCH 11.8-CU will result in an error.

nightsnack commented 20 hours ago

Yes. Because you use CUDA 11.8, and we use 12.4. Not that familiar with cuda and I don't know how to support CUDA 11.8.