| [ | |
| { | |
| "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)", | |
| "shortDescription" : "Same shape and dtype as the `sample` input. The predicted noise to faciliate the reverse diffusion (denoising) process", | |
| "shape" : "[]", | |
| "name" : "noise_pred", | |
| "type" : "MultiArray" | |
| } | |
| ], | |
| "version" : "CompVis\/stable-diffusion-v1-4", | |
| "modelParameters" : [ | |
| ], | |
| "author" : "Please refer to the Model Card available at huggingface.co\/CompVis\/stable-diffusion-v1-4", | |
| "specificationVersion" : 7, | |
| "storagePrecision" : "Float16", | |
| "license" : "OpenRAIL (https:\/\/huggingface.co\/spaces\/CompVis\/stable-diffusion-license)", | |
| "mlProgramOperationTypeHistogram" : { | |
| "Transpose" : 32, | |
| "UpsampleNearestNeighbor" : 3, | |
| "Ios16.reduceMean" : 218, | |
| "Ios16.sin" : 1, | |
| "Ios16.softmax" : 256, | |
| "Split" : 16, | |
| "Ios16.add" : 265, | |
| "Concat" : 46, | |
| "Ios16.realDiv" : 61, | |
| "Ios16.square" : 61, | |
| "ExpandDims" : 3, | |
| "Ios16.sub" : 109, | |
| "Ios16.cast" : 1, | |
| "Ios16.conv" : 282, | |
| "Ios16.einsum" : 512, | |
| "Ios16.gelu" : 16, | |
| "Ios16.reshape" : 154, | |
| "Ios16.batchNorm" : 61, | |
| "Ios16.rsqrt" : 48, | |
| "Ios16.silu" : 47, | |
| "Ios16.sqrt" : 61, | |
| "SliceByIndex" : 770, | |
| "Ios16.mul" : 417, | |
| "Ios16.cos" : 1 | |
| }, | |
| "computePrecision" : "Mixed (Float32, Float16, Int32)", | |
| "isUpdatable" : "0", | |
| "availability" : { | |
| "macOS" : "13.0", | |
| "tvOS" : "16.0", | |
| "watchOS" : "9.0", | |
| "iOS" : "16.0", | |
| "macCatalyst" : "16.0" | |
| }, | |
| "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.coremltools.version" : "6.1", | |
| "com.github.apple.coremltools.source" : "torch==1.13.0" | |
| }, | |
| "generatedClassName" : "Stable_Diffusion_version_CompVis_stable_diffusion_v1_4_unet", | |
| "method" : "predict" | |
| } | |
| ] |