Update VAE Encoder.

See https://github.com/apple/ml-stable-diffusion/issues/176

original/compiled/VAEEncoder.mlmodelc/analytics/coremldata.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e27fb7ebcffa0ff79c7212e1e914dbc14fddb8265b8c73d0d322a7eea4ef01e8
+oid sha256:e47b4f8e21d3662a8c0f77eb6c760543f6f08a83b4dc3fbc9d25aa1fa0f2ab66
 size 207

original/compiled/VAEEncoder.mlmodelc/coremldata.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b8cb326a2c2071e8ba0b1ef1084609061d0e86a603f55d066d25c60e04acd4a
-size 1272
+oid sha256:0896e8b3111c8950e88fbdb47c39b7c3bcee5fa5da8d3de5be91b8e6d567fc74
+size 759

original/compiled/VAEEncoder.mlmodelc/metadata.json

@@ -10,7 +10,7 @@
         "formattedType" : "MultiArray (Float32)",
         "shortDescription" : "The latent embeddings from the unet model from the input image.",
         "shape" : "[]",
-        "name" : "latent_dist",
+        "name" : "latent",
         "type" : "MultiArray"
       }
     ],
@@ -23,26 +23,23 @@
     "storagePrecision" : "Float16",
     "license" : "OpenRAIL (https:\/\/huggingface.co\/spaces\/CompVis\/stable-diffusion-license)",
     "mlProgramOperationTypeHistogram" : {
-      "Transpose" : 7,
-      "Ios16.exp" : 1,
-      "Ios16.reduceMean" : 44,
-      "Ios16.softmax" : 1,
-      "Split" : 1,
-      "Ios16.linear" : 4,
-      "Ios16.add" : 35,
-      "Ios16.realDiv" : 22,
-      "Ios16.square" : 22,
       "Pad" : 3,
-      "Ios16.sub" : 22,
       "Ios16.cast" : 1,
-      "Ios16.clip" : 1,
+      "Ios16.mul" : 1,
+      "Ios16.sqrt" : 22,
+      "Ios16.sub" : 22,
+      "Transpose" : 7,
       "Ios16.conv" : 28,
+      "Ios16.add" : 33,
+      "Ios16.linear" : 4,
       "Ios16.matmul" : 2,
-      "Ios16.reshape" : 54,
+      "Ios16.realDiv" : 22,
+      "Ios16.reduceMean" : 44,
+      "Ios16.softmax" : 1,
       "Ios16.batchNorm" : 22,
-      "Ios16.silu" : 21,
-      "Ios16.sqrt" : 22,
-      "Ios16.mul" : 6
+      "Ios16.square" : 22,
+      "Ios16.reshape" : 54,
+      "Ios16.silu" : 21
     },
     "computePrecision" : "Mixed (Float32, Float16, Int32)",
     "isUpdatable" : "0",
@@ -62,55 +59,15 @@
         "isOptional" : "0",
         "dataType" : "Float16",
         "formattedType" : "MultiArray (Float16 1 × 3 × 512 × 512)",
-        "shortDescription" : "An image of the correct size to create the latent space with, image2image and in-painting.",
+        "shortDescription" : "The input image to base the initial latents on normalized to range [-1, 1]",
         "shape" : "[1, 3, 512, 512]",
-        "name" : "sample",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 4 × 64 × 64)",
-        "shortDescription" : "Latent noise for `DiagonalGaussianDistribution` operation.",
-        "shape" : "[1, 4, 64, 64]",
-        "name" : "diagonal_noise",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 4 × 64 × 64)",
-        "shortDescription" : "Latent noise for use with strength parameter of image2image",
-        "shape" : "[1, 4, 64, 64]",
-        "name" : "noise",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 1)",
-        "shortDescription" : "Precalculated `sqrt_alphas_cumprod` value based on strength and the current schedular's alphasCumprod values",
-        "shape" : "[1, 1]",
-        "name" : "sqrt_alphas_cumprod",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 1)",
-        "shortDescription" : "Precalculated `sqrt_one_minus_alphas_cumprod` value based on strength and the current schedular's alphasCumprod values",
-        "shape" : "[1, 1]",
-        "name" : "sqrt_one_minus_alphas_cumprod",
+        "name" : "z",
         "type" : "MultiArray"
       }
     ],
     "userDefinedMetadata" : {
-      "com.github.apple.coremltools.version" : "6.1",
-      "com.github.apple.coremltools.source" : "torch==1.13.0"
+      "com.github.apple.coremltools.version" : "6.3.0",
+      "com.github.apple.coremltools.source" : "torch==2.0.0"
     },
     "generatedClassName" : "Stable_Diffusion_version_CompVis_stable_diffusion_v1_4_vae_encoder",
     "method" : "predict"

original/compiled/VAEEncoder.mlmodelc/model.mil

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:823a50e22d4989acd5247f33e8cc52e0caf5dbceec978fe76f6a5dd9759c2c7f
-size 137752
+oid sha256:d33d5d475d2457cb48abae239a2ba0a6ff14a6fa344633e4866c74c6f7907724
+size 135576

original/compiled/VAEEncoder.mlmodelc/weights/weight.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:edc9a37ae7f0c7b1696798514fa2224d183e5154356a356c0975acb6e9e7bba6
+oid sha256:527539d44d78d725c785dff135c1fb1529908f67069e5ed71bafd56e74a60493
 size 68370240

original/packages/Stable_Diffusion_version_CompVis_stable-diffusion-v1-4_vae_encoder.mlpackage/Data/com.apple.CoreML/model.mlmodel

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6269f295fe1eef3d3194186afbf56d932b399d528663ff95fc56cfdf1885d015
-size 117448
+oid sha256:e2b68542ca85526d3d48e17f04ac406b9c2f5879dfa54fe39c7981bb41ab7cd9
+size 114902

original/packages/Stable_Diffusion_version_CompVis_stable-diffusion-v1-4_vae_encoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:edc9a37ae7f0c7b1696798514fa2224d183e5154356a356c0975acb6e9e7bba6
+oid sha256:527539d44d78d725c785dff135c1fb1529908f67069e5ed71bafd56e74a60493
 size 68370240

original/packages/Stable_Diffusion_version_CompVis_stable-diffusion-v1-4_vae_encoder.mlpackage/Manifest.json

@@ -1,18 +1,18 @@
 {
     "fileFormatVersion": "1.0.0",
     "itemInfoEntries": {
-        "572B88FB-BD8C-4BB2-94A9-59BBAA98B7E6": {
-            "author": "com.apple.CoreML",
-            "description": "CoreML Model Weights",
-            "name": "weights",
-            "path": "com.apple.CoreML/weights"
-        },
-        "908BDF53-CD61-4E50-821B-0721632446FB": {
+        "4AC76ED5-4FE2-4025-A4F9-190A8B2EFE35": {
             "author": "com.apple.CoreML",
             "description": "CoreML Model Specification",
             "name": "model.mlmodel",
             "path": "com.apple.CoreML/model.mlmodel"
+        },
+        "E3770AC4-12AE-42B0-8181-C0F84452E3D3": {
+            "author": "com.apple.CoreML",
+            "description": "CoreML Model Weights",
+            "name": "weights",
+            "path": "com.apple.CoreML/weights"
         }
     },
-    "rootModelIdentifier": "908BDF53-CD61-4E50-821B-0721632446FB"
+    "rootModelIdentifier": "4AC76ED5-4FE2-4025-A4F9-190A8B2EFE35"
 }

split_einsum/compiled/VAEEncoder.mlmodelc/analytics/coremldata.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e27fb7ebcffa0ff79c7212e1e914dbc14fddb8265b8c73d0d322a7eea4ef01e8
+oid sha256:e47b4f8e21d3662a8c0f77eb6c760543f6f08a83b4dc3fbc9d25aa1fa0f2ab66
 size 207

split_einsum/compiled/VAEEncoder.mlmodelc/coremldata.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a8053f3c4f3cebb592d42dd6e2dad8758c4c40417c8552424faf74c9986fe3f
-size 1272
+oid sha256:0896e8b3111c8950e88fbdb47c39b7c3bcee5fa5da8d3de5be91b8e6d567fc74
+size 759

split_einsum/compiled/VAEEncoder.mlmodelc/metadata.json

@@ -10,7 +10,7 @@
         "formattedType" : "MultiArray (Float32)",
         "shortDescription" : "The latent embeddings from the unet model from the input image.",
         "shape" : "[]",
-        "name" : "latent_dist",
+        "name" : "latent",
         "type" : "MultiArray"
       }
     ],
@@ -23,26 +23,23 @@
     "storagePrecision" : "Float16",
     "license" : "OpenRAIL (https:\/\/huggingface.co\/spaces\/CompVis\/stable-diffusion-license)",
     "mlProgramOperationTypeHistogram" : {
-      "Transpose" : 7,
-      "Ios16.exp" : 1,
-      "Ios16.reduceMean" : 44,
-      "Ios16.softmax" : 1,
-      "Split" : 1,
-      "Ios16.linear" : 4,
-      "Ios16.add" : 35,
-      "Ios16.realDiv" : 22,
-      "Ios16.square" : 22,
       "Pad" : 3,
-      "Ios16.sub" : 22,
       "Ios16.cast" : 1,
-      "Ios16.clip" : 1,
+      "Ios16.mul" : 1,
+      "Ios16.sqrt" : 22,
+      "Ios16.sub" : 22,
+      "Transpose" : 7,
       "Ios16.conv" : 28,
+      "Ios16.add" : 33,
+      "Ios16.linear" : 4,
       "Ios16.matmul" : 2,
-      "Ios16.reshape" : 54,
+      "Ios16.realDiv" : 22,
+      "Ios16.reduceMean" : 44,
+      "Ios16.softmax" : 1,
       "Ios16.batchNorm" : 22,
-      "Ios16.silu" : 21,
-      "Ios16.sqrt" : 22,
-      "Ios16.mul" : 6
+      "Ios16.square" : 22,
+      "Ios16.reshape" : 54,
+      "Ios16.silu" : 21
     },
     "computePrecision" : "Mixed (Float32, Float16, Int32)",
     "isUpdatable" : "0",
@@ -62,55 +59,15 @@
         "isOptional" : "0",
         "dataType" : "Float16",
         "formattedType" : "MultiArray (Float16 1 × 3 × 512 × 512)",
-        "shortDescription" : "An image of the correct size to create the latent space with, image2image and in-painting.",
+        "shortDescription" : "The input image to base the initial latents on normalized to range [-1, 1]",
         "shape" : "[1, 3, 512, 512]",
-        "name" : "sample",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 4 × 64 × 64)",
-        "shortDescription" : "Latent noise for `DiagonalGaussianDistribution` operation.",
-        "shape" : "[1, 4, 64, 64]",
-        "name" : "diagonal_noise",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 4 × 64 × 64)",
-        "shortDescription" : "Latent noise for use with strength parameter of image2image",
-        "shape" : "[1, 4, 64, 64]",
-        "name" : "noise",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 1)",
-        "shortDescription" : "Precalculated `sqrt_alphas_cumprod` value based on strength and the current schedular's alphasCumprod values",
-        "shape" : "[1, 1]",
-        "name" : "sqrt_alphas_cumprod",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 1)",
-        "shortDescription" : "Precalculated `sqrt_one_minus_alphas_cumprod` value based on strength and the current schedular's alphasCumprod values",
-        "shape" : "[1, 1]",
-        "name" : "sqrt_one_minus_alphas_cumprod",
+        "name" : "z",
         "type" : "MultiArray"
       }
     ],
     "userDefinedMetadata" : {
-      "com.github.apple.coremltools.version" : "6.1",
-      "com.github.apple.coremltools.source" : "torch==1.13.0"
+      "com.github.apple.coremltools.version" : "6.3.0",
+      "com.github.apple.coremltools.source" : "torch==2.0.0"
     },
     "generatedClassName" : "Stable_Diffusion_version_CompVis_stable_diffusion_v1_4_vae_encoder",
     "method" : "predict"

split_einsum/compiled/VAEEncoder.mlmodelc/model.mil

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:823a50e22d4989acd5247f33e8cc52e0caf5dbceec978fe76f6a5dd9759c2c7f
-size 137752
+oid sha256:d33d5d475d2457cb48abae239a2ba0a6ff14a6fa344633e4866c74c6f7907724
+size 135576

split_einsum/compiled/VAEEncoder.mlmodelc/weights/weight.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2604aac03e12dde2d56b35a7d2f6c3275860b049ea2dedd75426c2a5c3a6c667
+oid sha256:527539d44d78d725c785dff135c1fb1529908f67069e5ed71bafd56e74a60493
 size 68370240

split_einsum/packages/Stable_Diffusion_version_CompVis_stable-diffusion-v1-4_vae_encoder.mlpackage/Data/com.apple.CoreML/model.mlmodel

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6269f295fe1eef3d3194186afbf56d932b399d528663ff95fc56cfdf1885d015
-size 117448
+oid sha256:e2b68542ca85526d3d48e17f04ac406b9c2f5879dfa54fe39c7981bb41ab7cd9
+size 114902

split_einsum/packages/Stable_Diffusion_version_CompVis_stable-diffusion-v1-4_vae_encoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2604aac03e12dde2d56b35a7d2f6c3275860b049ea2dedd75426c2a5c3a6c667
+oid sha256:527539d44d78d725c785dff135c1fb1529908f67069e5ed71bafd56e74a60493
 size 68370240

split_einsum/packages/Stable_Diffusion_version_CompVis_stable-diffusion-v1-4_vae_encoder.mlpackage/Manifest.json

@@ -1,18 +1,18 @@
 {
     "fileFormatVersion": "1.0.0",
     "itemInfoEntries": {
-        "73F723E9-BBE9-40B7-8113-8FCA8733A20D": {
+        "4AC76ED5-4FE2-4025-A4F9-190A8B2EFE35": {
             "author": "com.apple.CoreML",
             "description": "CoreML Model Specification",
             "name": "model.mlmodel",
             "path": "com.apple.CoreML/model.mlmodel"
         },
-        "DEB5F935-76BB-4FD1-BBD6-968993E96111": {
+        "E3770AC4-12AE-42B0-8181-C0F84452E3D3": {
             "author": "com.apple.CoreML",
             "description": "CoreML Model Weights",
             "name": "weights",
             "path": "com.apple.CoreML/weights"
         }
     },
-    "rootModelIdentifier": "73F723E9-BBE9-40B7-8113-8FCA8733A20D"
+    "rootModelIdentifier": "4AC76ED5-4FE2-4025-A4F9-190A8B2EFE35"
 }