[
  {
    "shortDescription" : "Stable Diffusion generates images conditioned on text or other images as input through the diffusion process. Please refer to https:\/\/arxiv.org\/abs\/2112.10752 for details.",
    "metadataOutputVersion" : "3.0",
    "outputSchema" : [
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float32",
        "formattedType" : "MultiArray (Float32 2 × 4 × 64 × 64)",
        "shortDescription" : "Same shape and dtype as the `sample` input. The predicted noise to facilitate the reverse diffusion (denoising) process",
        "shape" : "[2, 4, 64, 64]",
        "name" : "noise_pred",
        "type" : "MultiArray"
      }
    ],
    "version" : "cyberdelia\/CyberRealistic@99827f96edd717dacb28c68560680c201c55df05 + latent-consistency\/lcm-lora-sdv1-5@cf2fced511dbe7e26c8d1d397e728fbab875db4b",
    "modelParameters" : [

    ],
    "author" : "Cyberdelia + Latent Consistency Models; Core ML 6-bit palettized conversion prepared for LocalMuse",
    "specificationVersion" : 8,
    "storagePrecision" : "Mixed (Float16, Palettized (6 bits))",
    "license" : "See the original model terms at https:\/\/huggingface.co\/spaces\/CompVis\/stable-diffusion-license",
    "mlProgramOperationTypeHistogram" : {
      "Ios17.square" : 61,
      "Ios17.cos" : 1,
      "UpsampleNearestNeighbor" : 3,
      "Ios17.reshape" : 154,
      "Ios16.reduceMean" : 122,
      "Ios16.softmax" : 896,
      "Ios17.sin" : 1,
      "Ios17.transpose" : 32,
      "Split" : 16,
      "Ios17.expandDims" : 3,
      "Ios17.add" : 169,
      "Ios17.sliceByIndex" : 1570,
      "Ios17.layerNorm" : 48,
      "Ios17.batchNorm" : 61,
      "Ios17.sqrt" : 61,
      "Ios17.sub" : 61,
      "Ios17.conv" : 282,
      "Ios16.constexprLutToDense" : 282,
      "Ios17.realDiv" : 61,
      "Ios17.concat" : 206,
      "Ios16.einsum" : 1792,
      "Ios16.silu" : 47,
      "Ios16.gelu" : 16,
      "Ios17.cast" : 1,
      "Ios17.mul" : 913
    },
    "computePrecision" : "Mixed (Float16, Float32, Int32)",
    "stateSchema" : [

    ],
    "isUpdatable" : "0",
    "availability" : {
      "macOS" : "14.0",
      "tvOS" : "17.0",
      "visionOS" : "1.0",
      "watchOS" : "10.0",
      "iOS" : "17.0",
      "macCatalyst" : "17.0"
    },
    "modelType" : {
      "name" : "MLModelType_mlProgram"
    },
    "inputSchema" : [
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 2 × 4 × 64 × 64)",
        "shortDescription" : "The low resolution latent feature maps being denoised through reverse diffusion",
        "shape" : "[2, 4, 64, 64]",
        "name" : "sample",
        "type" : "MultiArray"
      },
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 2)",
        "shortDescription" : "A value emitted by the associated scheduler object to condition the model on a given noise schedule",
        "shape" : "[2]",
        "name" : "timestep",
        "type" : "MultiArray"
      },
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 2 × 768 × 1 × 77)",
        "shortDescription" : "Output embeddings from the associated text_encoder model to condition to generated image on text. A maximum of 77 tokens (~40 words) are allowed. Longer text is truncated. Shorter text does not reduce computation.",
        "shape" : "[2, 768, 1, 77]",
        "name" : "encoder_hidden_states",
        "type" : "MultiArray"
      }
    ],
    "userDefinedMetadata" : {
      "com.github.apple.ml-stable-diffusion.version" : "1.1.0",
      "com.github.apple.coremltools.source" : "torch==2.4.0",
      "com.github.apple.coremltools.version" : "8.3.0",
      "com.github.apple.coremltools.source_dialect" : "TorchScript"
    },
    "generatedClassName" : "Stable_Diffusion_version_cyberdelia_CyberRealistic_unet",
    "method" : "predict"
  }
]