| { |
| "_class_name": "FoleyOmniConfig", |
| "_name_or_path": "CocoBro/Foley-Omni", |
| "architectures": [ |
| "FoleyOmniDiT" |
| ], |
| "model_type": "t2a", |
| "patch_size": [1], |
| "dim": 3072, |
| "ffn_dim": 14336, |
| "freq_dim": 256, |
| "num_heads": 24, |
| "num_layers": 30, |
| "in_dim": 20, |
| "out_dim": 20, |
| "text_len": 512, |
| "window_size": [-1, -1], |
| "qk_norm": true, |
| "cross_attn_norm": true, |
| "eps": 1e-6, |
| "temporal_rope_scaling_factor": 0.19676, |
| "sync_drop_from_context": false, |
| "sync_add_to_audio": true, |
| "sync_add_to_audio_each_block": true, |
| "sync_use_adaln": false, |
| "sync_use_proj": true, |
| "sync_use_adapter": true, |
| "sync_adapter_type": "depthwise_conv", |
| "sync_adapter_kernel_size": 5, |
| "sync_adapter_expansion": 2.0, |
| "use_continuous_time_context_rope": false, |
| "continuous_time_context_rope_mode": "auto", |
| "use_text_context_rope": true, |
| "use_clip_context_rope": true, |
| "use_sync_context_rope": true, |
| "context_rope_theta": 10000.0, |
| "context_text_rope_scaling": 1.0, |
| "context_clip_rope_scaling": 1.0, |
| "context_sync_rope_scaling": 1.0 |
| } |
|
|
