{
  "model_type": "vimogen",
  "checkpoint": "model.pt",
  "mean_path": "assets/meta/mean.npy",
  "std_path": "assets/meta/std.npy",
  "motion_representation": "vimogen276",
  "cfg_scale": 5.0,
  "denoising_strength": 0.7,
  "num_inference_steps": 50,
  "text_encoder": "Wan2.1-T2V-1.3B UMT5-XXL",
  "source": "ViMoGen released HumanML3D checkpoint",
  "_class_name": "ViMoGenPipeline",
  "_library_name": "motius",
  "format_version": 1,
  "pipeline_class": "motius.pipelines.vimogen.ViMoGenPipeline",
  "bundle_class": "motius.models.vimogen.ViMoGenBundle",
  "tasks": [
    "text_to_motion"
  ],
  "required_files": [
    "assets/meta/mean.npy",
    "assets/meta/std.npy",
    "context_null_padded.pth",
    "model.pt",
    "wan/config.json",
    "wan/google/umt5-xxl/special_tokens_map.json",
    "wan/google/umt5-xxl/spiece.model",
    "wan/google/umt5-xxl/tokenizer.json",
    "wan/google/umt5-xxl/tokenizer_config.json",
    "wan/models_t5_umt5-xxl-enc-bf16.pth"
  ],
  "api": {
    "loader": "motius.Pipeline.from_pretrained",
    "task_methods": [
      "infer_text_to_motion"
    ]
  },
  "artifact_format": "motius-vimogen-v1",
  "wan_dir": "wan",
  "text_encoder_stored_in_artifact": true,
  "artifacts": {
    "wan_text_encoder": "wan"
  },
  "components": {
    "text_encoder": {
      "stored_in_artifact": true,
      "path": "wan"
    }
  }
}