Instructions to use utkarshUY/CharacterPlex-FullFT-v4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Moshi
How to use utkarshUY/CharacterPlex-FullFT-v4 with Moshi:
# pip install moshi # Run the interactive web server python -m moshi.server --hf-repo "utkarshUY/CharacterPlex-FullFT-v4" # Then open https://localhost:8998 in your browser
# pip install moshi import torch from moshi.models import loaders # Load checkpoint info from HuggingFace checkpoint = loaders.CheckpointInfo.from_hf_repo("utkarshUY/CharacterPlex-FullFT-v4") # Load the Mimi audio codec mimi = checkpoint.get_mimi(device="cuda") mimi.set_num_codebooks(8) # Encode audio (24kHz, mono) wav = torch.randn(1, 1, 24000 * 10) # [batch, channels, samples] with torch.no_grad(): codes = mimi.encode(wav.cuda()) decoded = mimi.decode(codes) - Notebooks
- Google Colab
- Kaggle
Download memory.json from utkarshUY/CharacterPlex-FullFT-v4: direct link, hf CLI and curl.
- Browser
- Download file 179 Bytes
-
https://huggingface.co/utkarshUY/CharacterPlex-FullFT-v4/resolve/main/memory.json
- Command line
-
hf download hf://utkarshUY/CharacterPlex-FullFT-v4/memory.json
-
curl -L -o memory.json https://huggingface.co/utkarshUY/CharacterPlex-FullFT-v4/resolve/main/memory.json
179 Bytes
| { | |
| "notes_frames": 64, | |
| "silence_frames": 6, | |
| "window": null, | |
| "directive_lut": 0, | |
| "directive_classes": [], | |
| "extra_slots": 0, | |
| "memory_trained": true, | |
| "card_max_tokens": 1000 | |
| } |