| [ |
| { |
| "metadataOutputVersion" : "3.0", |
| "storagePrecision" : "Mixed (Float16, Float32)", |
| "outputSchema" : [ |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 2 × 1 × 512 × 12 × 64)", |
| "shortDescription" : "", |
| "shape" : "[2, 1, 512, 12, 64]", |
| "name" : "var_208", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 2 × 1 × 512 × 12 × 64)", |
| "shortDescription" : "", |
| "shape" : "[2, 1, 512, 12, 64]", |
| "name" : "var_374", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 2 × 1 × 512 × 12 × 64)", |
| "shortDescription" : "", |
| "shape" : "[2, 1, 512, 12, 64]", |
| "name" : "var_540", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 2 × 1 × 512 × 12 × 64)", |
| "shortDescription" : "", |
| "shape" : "[2, 1, 512, 12, 64]", |
| "name" : "var_706", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 2 × 1 × 512 × 12 × 64)", |
| "shortDescription" : "", |
| "shape" : "[2, 1, 512, 12, 64]", |
| "name" : "var_872", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 2 × 1 × 512 × 12 × 64)", |
| "shortDescription" : "", |
| "shape" : "[2, 1, 512, 12, 64]", |
| "name" : "var_1038", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 2 × 1 × 512 × 12 × 64)", |
| "shortDescription" : "", |
| "shape" : "[2, 1, 512, 12, 64]", |
| "name" : "var_1204", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 2 × 1 × 512 × 12 × 64)", |
| "shortDescription" : "", |
| "shape" : "[2, 1, 512, 12, 64]", |
| "name" : "var_1370", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 2 × 1 × 512 × 12 × 64)", |
| "shortDescription" : "", |
| "shape" : "[2, 1, 512, 12, 64]", |
| "name" : "var_1536", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 2 × 1 × 512 × 12 × 64)", |
| "shortDescription" : "", |
| "shape" : "[2, 1, 512, 12, 64]", |
| "name" : "var_1702", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 2 × 1 × 512 × 12 × 64)", |
| "shortDescription" : "", |
| "shape" : "[2, 1, 512, 12, 64]", |
| "name" : "var_1868", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 2 × 1 × 512 × 12 × 64)", |
| "shortDescription" : "", |
| "shape" : "[2, 1, 512, 12, 64]", |
| "name" : "var_1958", |
| "type" : "MultiArray" |
| } |
| ], |
| "modelParameters" : [ |
|
|
| ], |
| "specificationVersion" : 8, |
| "mlProgramOperationTypeHistogram" : { |
| "Pad" : 24, |
| "Ios17.mul" : 22, |
| "Ios17.linear" : 56, |
| "Ios17.transpose" : 110, |
| "Select" : 11, |
| "Ios17.matmul" : 44, |
| "Ios17.conv" : 22, |
| "Stack" : 12, |
| "Ios17.sliceByIndex" : 57, |
| "Ios17.add" : 45, |
| "Ios16.softmax" : 22, |
| "Ios17.expandDims" : 2, |
| "Ios16.gelu" : 11, |
| "Ios17.logicalNot" : 1, |
| "Ios17.reshape" : 56, |
| "Ios17.layerNorm" : 45, |
| "Ios17.cast" : 23 |
| }, |
| "computePrecision" : "Mixed (Float16, Float32, Int32)", |
| "isUpdatable" : "0", |
| "stateSchema" : [ |
|
|
| ], |
| "availability" : { |
| "macOS" : "14.0", |
| "tvOS" : "17.0", |
| "visionOS" : "1.0", |
| "watchOS" : "10.0", |
| "iOS" : "17.0", |
| "macCatalyst" : "17.0" |
| }, |
| "modelType" : { |
| "name" : "MLModelType_mlProgram" |
| }, |
| "userDefinedMetadata" : { |
| "com.github.apple.coremltools.conversion_date" : "2026-04-24", |
| "com.github.apple.coremltools.source" : "torch==2.10.0", |
| "com.github.apple.coremltools.version" : "9.0", |
| "com.github.apple.coremltools.source_dialect" : "TorchScript" |
| }, |
| "inputSchema" : [ |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 1 × 110 × 768)", |
| "shortDescription" : "", |
| "shape" : "[1, 110, 768]", |
| "name" : "audio_embed", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float16", |
| "formattedType" : "MultiArray (Float16 1 × 256 × 768)", |
| "shortDescription" : "", |
| "shape" : "[1, 256, 768]", |
| "name" : "encoder_output", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float32", |
| "formattedType" : "MultiArray (Float32 1 × 256)", |
| "shortDescription" : "", |
| "shape" : "[1, 256]", |
| "name" : "encoder_mask", |
| "type" : "MultiArray" |
| } |
| ], |
| "generatedClassName" : "decoder_prefill", |
| "method" : "predict" |
| } |
| ] |