{
  "family": "triposg",
  "image_encoder": {
    "_name_or_path": "facebook/dinov2-large",
    "apply_layernorm": true,
    "architectures": [
      "Dinov2Model"
    ],
    "attention_probs_dropout_prob": 0.0,
    "drop_path_rate": 0.0,
    "hidden_act": "gelu",
    "hidden_dropout_prob": 0.0,
    "hidden_size": 1024,
    "image_size": 518,
    "initializer_range": 0.02,
    "layer_norm_eps": 1e-06,
    "layerscale_value": 1.0,
    "mlp_ratio": 4,
    "model_type": "dinov2",
    "num_attention_heads": 16,
    "num_channels": 3,
    "num_hidden_layers": 24,
    "out_features": [
      "stage24"
    ],
    "out_indices": [
      24
    ],
    "patch_size": 14,
    "qkv_bias": true,
    "reshape_hidden_states": true,
    "stage_names": [
      "stem",
      "stage1",
      "stage2",
      "stage3",
      "stage4",
      "stage5",
      "stage6",
      "stage7",
      "stage8",
      "stage9",
      "stage10",
      "stage11",
      "stage12",
      "stage13",
      "stage14",
      "stage15",
      "stage16",
      "stage17",
      "stage18",
      "stage19",
      "stage20",
      "stage21",
      "stage22",
      "stage23",
      "stage24"
    ],
    "torch_dtype": "float32",
    "transformers_version": "4.45.2",
    "use_swiglu_ffn": false
  },
  "model_index": {
    "_class_name": "TripoSGPipeline",
    "_diffusers_version": "0.30.3",
    "feature_extractor_dinov2": [
      "transformers",
      "BitImageProcessor"
    ],
    "image_encoder_dinov2": [
      "transformers",
      "Dinov2Model"
    ],
    "scheduler": [
      "triposg.schedulers.scheduling_rectified_flow",
      "RectifiedFlowScheduler"
    ],
    "transformer": [
      "triposg.models.transformers.triposg_transformer",
      "TripoSGDiTModel"
    ],
    "vae": [
      "triposg.models.autoencoders.autoencoder_kl_triposg",
      "TripoSGVAEModel"
    ]
  },
  "pipeline_defaults": {
    "bounds": [
      -1.005,
      -1.005,
      -1.005,
      1.005,
      1.005,
      1.005
    ],
    "dense_octree_depth": 8,
    "guidance_scale": 7.0,
    "hierarchical_octree_depth": 9,
    "num_inference_steps": 50,
    "num_tokens": 2048,
    "sdf_level": 0.0
  },
  "preprocessor": {
    "crop_size": {
      "height": 224,
      "width": 224
    },
    "do_center_crop": true,
    "do_convert_rgb": true,
    "do_normalize": true,
    "do_rescale": true,
    "do_resize": true,
    "image_mean": [
      0.485,
      0.456,
      0.406
    ],
    "image_processor_type": "BitImageProcessor",
    "image_std": [
      0.229,
      0.224,
      0.225
    ],
    "resample": 3,
    "rescale_factor": 0.00392156862745098,
    "size": {
      "shortest_edge": 256
    }
  },
  "scheduler_config": {
    "_class_name": "RectifiedFlowScheduler",
    "_diffusers_version": "0.30.3",
    "num_train_timesteps": 1000,
    "shift": 1,
    "use_dynamic_shifting": false
  },
  "transformer": {
    "_class_name": "TripoSGDiTModel",
    "_diffusers_version": "0.30.3",
    "cross_attention_dim": 1024,
    "in_channels": 64,
    "num_attention_heads": 16,
    "num_layers": 21,
    "width": 2048
  },
  "vae": {
    "_class_name": "Tripo2VAEModel",
    "_diffusers_version": "0.30.3",
    "embed_frequency": 8,
    "embed_include_pi": false,
    "embedding_type": "frequency",
    "in_channels": 3,
    "latent_channels": 64,
    "num_attention_heads": 8,
    "num_layers_decoder": 16,
    "num_layers_encoder": 8,
    "width_decoder": 1024,
    "width_encoder": 512
  }
}