Views
No views yet
1ckpts/
2├── Foley-Omni/
3│ └── v2st.pth
4├── Wan2.2-TI2V-5B/
5│ ├── models_t5_umt5-xxl-enc-bf16.pth
6│ └── google/
7│ └── umt5-xxl/
8│ ├── special_tokens_map.json
9│ ├── spiece.model
10│ ├── tokenizer.json
11│ └── tokenizer_config.json
12└── mmaudio/
13 └── ext_weights/
14 ├── v1-16.pth
15 ├── best_netG.pt
16 └── synchformer_state_dict.pthckpts/Foley-Omni/v2st.pth: released inference-only Foley-Omni weightsckpts/Wan2.2-TI2V-5B/*: text encoder and tokenizer for text conditioningckpts/mmaudio/ext_weights/v1-16.pth: audio VAE for the 16 kHz inference pathckpts/mmaudio/ext_weights/best_netG.pt: vocoder for waveform decodingckpts/mmaudio/ext_weights/synchformer_state_dict.pth: online visual feature extractionclip_feature_path and sync_feature_pathsynchformer_state_dict.pth is included in this repository because it is required for online Sync feature extraction.open_clip from apple/DFN5B-CLIP-ViT-H-14-384 on first use. The current code path does not use a separate local CLIP checkpoint file.