Text-To-Video demo
Set chunk_size to 2, merging_ratio to 1.0, and the text-2-video model can be run with RTX 3080.
Text-To-Video with Pose Control demo
Failed to run Text-To-Video with Pose Control demo on RTX 3080, even set chunk_size to 1, merging_ratio to 1.0 and save_memory to True in config.py.
OutOfMemoryError: CUDA out of memory. Tried to allocate 4.00 GiB (GPU 0; 9.77 GiB total capacity; 3.90 GiB already allocated; 2.71 GiB free; 5.21 GiB reserved in total by PyTorch) If reserved memory is >>
allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
demo
Device NVIDIA GeForce RTX 3080, 10 GB GPU Memory
Text-To-Video demo Set
chunk_sizeto 2,merging_ratioto 1.0, and the text-2-video model can be run with RTX 3080.Text-To-Video with Pose Control demo Failed to run Text-To-Video with Pose Control demo on RTX 3080, even set
chunk_sizeto 1,merging_ratioto 1.0 andsave_memoryto True in config.py.