{
  "format": "ambit",
  "weights": "model.safetensors",
  "vae_weights": "vae.safetensors",
  "renderer_config": "renderer.json",
  "vae_config": "vae.json",
  "codec": "codec",
  "text_encoder": "text_encoder",
  "planner_encoder": "planner_encoder",
  "clap": {
    "sample_rate": 44100,
    "vae_hop": 1024,
    "latent_channels": 64,
    "max_frames": 648,
    "stride": 4,
    "width": 384,
    "layers": 6,
    "heads": 6,
    "semantic_dim": 512,
    "scene_dim": 256,
    "text_dim": 1024,
    "dropout": 0.1
  },
  "structure": {},
  "transcript": {
    "output_dim": 768,
    "width": 256,
    "blocks": 4,
    "max_bytes": 4096
  },
  "audio_transcript": {
    "width": 256,
    "heads": 4
  },
  "planner": {
    "hidden_size": 2560,
    "head_blocks": 4
  },
  "sampling": {
    "generation": {
      "steps": 100,
      "cfg_scale": 3.0
    },
    "editing": {
      "steps": 20,
      "cfg_scale": 1.0
    }
  }
}