Adds new models

ggml-base-encoder.mlmodelc.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e6ab77041942572f239b5b602f8aaa1c3ed29d73e3d8f20abea03a773541089
+oid sha256:9cf9278264a9e0f56a6649ca10ab2cde81f6bfe86133a1f08de15006c269cd39
 size 37922638

ggml-large-encoder.mlmodelc.zip → ggml-base-encoder.mlmodelc/analytics/coremldata.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c65d28737f51d09dff3b9da9d26c2ab6f5d82857c9b049ce383d64c2502a5541
-size 1174643458
+oid sha256:2441ae34fc7d12946dba7b63379063e856ffc7c3e11ba5f7533efb1450562ca6
+size 207

ggml-large.bin → ggml-base-encoder.mlmodelc/coremldata.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a423fe4d40c82774b6af34115b8b935f34152246eb19e80e376071d3f999487
-size 3094623691
+oid sha256:05fe28591b40616fa0c34ad7b853133623f5300923ec812acb11459c411acf3b
+size 149

ggml-base-encoder.mlmodelc/metadata.json

@@ -0,0 +1,64 @@
+[
+  {
+    "metadataOutputVersion" : "3.0",
+    "storagePrecision" : "Float16",
+    "outputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32)",
+        "shortDescription" : "",
+        "shape" : "[]",
+        "name" : "output",
+        "type" : "MultiArray"
+      }
+    ],
+    "modelParameters" : [
+
+    ],
+    "specificationVersion" : 6,
+    "mlProgramOperationTypeHistogram" : {
+      "Linear" : 36,
+      "Matmul" : 12,
+      "Cast" : 2,
+      "Conv" : 2,
+      "Softmax" : 6,
+      "Add" : 13,
+      "LayerNorm" : 13,
+      "Mul" : 12,
+      "Transpose" : 25,
+      "Gelu" : 8,
+      "Reshape" : 24
+    },
+    "computePrecision" : "Mixed (Float16, Float32, Int32)",
+    "isUpdatable" : "0",
+    "availability" : {
+      "macOS" : "12.0",
+      "tvOS" : "15.0",
+      "watchOS" : "8.0",
+      "iOS" : "15.0",
+      "macCatalyst" : "15.0"
+    },
+    "modelType" : {
+      "name" : "MLModelType_mlProgram"
+    },
+    "userDefinedMetadata" : {
+
+    },
+    "inputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32 1 × 80 × 3000)",
+        "shortDescription" : "",
+        "shape" : "[1, 80, 3000]",
+        "name" : "logmel_data",
+        "type" : "MultiArray"
+      }
+    ],
+    "generatedClassName" : "coreml_encoder_base",
+    "method" : "predict"
+  }
+]

ggml-base-encoder.mlmodelc/model.mil

@@ -0,0 +1,393 @@
+program(1.0)
+[buildInfo = dict<tensor<string, []>, tensor<string, []>>({{"coremlc-component-MIL", "4.28.4"}, {"coremlc-version", "1436.100.10"}})]
+{
+    func main<ios15>(tensor<fp32, [1, 80, 3000]> logmel_data) {
+            tensor<int32, []> var_20 = const()[name = tensor<string, []>("op_20"), val = tensor<int32, []>(1)];
+            tensor<int32, [1]> var_28 = const()[name = tensor<string, []>("op_28"), val = tensor<int32, [1]>([1])];
+            tensor<int32, [1]> var_30 = const()[name = tensor<string, []>("op_30"), val = tensor<int32, [1]>([1])];
+            tensor<string, []> var_32_pad_type_0 = const()[name = tensor<string, []>("op_32_pad_type_0"), val = tensor<string, []>("custom")];
+            tensor<int32, [2]> var_32_pad_0 = const()[name = tensor<string, []>("op_32_pad_0"), val = tensor<int32, [2]>([1, 1])];
+            tensor<string, []> logmel_data_to_fp16_dtype_0 = const()[name = tensor<string, []>("logmel_data_to_fp16_dtype_0"), val = tensor<string, []>("fp16")];
+            tensor<fp16, [512, 80, 3]> weight_3_to_fp16 = const()[name = tensor<string, []>("weight_3_to_fp16"), val = tensor<fp16, [512, 80, 3]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(64)))];
+            tensor<fp16, [512]> bias_3_to_fp16 = const()[name = tensor<string, []>("bias_3_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(245888)))];
+            tensor<fp16, [1, 80, 3000]> cast_187 = cast(dtype = logmel_data_to_fp16_dtype_0, x = logmel_data);
+            tensor<fp16, [1, 512, 3000]> var_32_cast = conv(bias = bias_3_to_fp16, dilations = var_30, groups = var_20, pad = var_32_pad_0, pad_type = var_32_pad_type_0, strides = var_28, weight = weight_3_to_fp16, x = cast_187);
+            tensor<string, []> input_1_mode_0 = const()[name = tensor<string, []>("input_1_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 512, 3000]> input_1_cast = gelu(mode = input_1_mode_0, x = var_32_cast);
+            tensor<int32, []> var_36 = const()[name = tensor<string, []>("op_36"), val = tensor<int32, []>(1)];
+            tensor<int32, [1]> var_45 = const()[name = tensor<string, []>("op_45"), val = tensor<int32, [1]>([2])];
+            tensor<int32, [1]> var_47 = const()[name = tensor<string, []>("op_47"), val = tensor<int32, [1]>([1])];
+            tensor<string, []> var_49_pad_type_0 = const()[name = tensor<string, []>("op_49_pad_type_0"), val = tensor<string, []>("custom")];
+            tensor<int32, [2]> var_49_pad_0 = const()[name = tensor<string, []>("op_49_pad_0"), val = tensor<int32, [2]>([1, 1])];
+            tensor<fp16, [512, 512, 3]> weight_7_to_fp16 = const()[name = tensor<string, []>("weight_7_to_fp16"), val = tensor<fp16, [512, 512, 3]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(246976)))];
+            tensor<fp16, [512]> bias_7_to_fp16 = const()[name = tensor<string, []>("bias_7_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1819904)))];
+            tensor<fp16, [1, 512, 1500]> var_49_cast = conv(bias = bias_7_to_fp16, dilations = var_47, groups = var_36, pad = var_49_pad_0, pad_type = var_49_pad_type_0, strides = var_45, weight = weight_7_to_fp16, x = input_1_cast);
+            tensor<string, []> x_3_mode_0 = const()[name = tensor<string, []>("x_3_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 512, 1500]> x_3_cast = gelu(mode = x_3_mode_0, x = var_49_cast);
+            tensor<int32, [3]> var_54 = const()[name = tensor<string, []>("op_54"), val = tensor<int32, [3]>([0, 2, 1])];
+            tensor<fp16, [1500, 512]> positional_embedding_to_fp16 = const()[name = tensor<string, []>("positional_embedding_to_fp16"), val = tensor<fp16, [1500, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1820992)))];
+            tensor<fp16, [1, 1500, 512]> transpose_48 = transpose(perm = var_54, x = x_3_cast);
+            tensor<fp16, [1, 1500, 512]> var_57_cast = add(x = transpose_48, y = positional_embedding_to_fp16);
+            tensor<int32, []> var_70 = const()[name = tensor<string, []>("op_70"), val = tensor<int32, []>(-1)];
+            tensor<int32, [1]> var_87_axes_0 = const()[name = tensor<string, []>("op_87_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_0_attn_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_0_attn_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3357056)))];
+            tensor<fp16, [512]> blocks_0_attn_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_0_attn_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3358144)))];
+            tensor<fp16, []> var_76_to_fp16 = const()[name = tensor<string, []>("op_76_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_87_cast = layer_norm(axes = var_87_axes_0, beta = blocks_0_attn_ln_bias_to_fp16, epsilon = var_76_to_fp16, gamma = blocks_0_attn_ln_weight_to_fp16, x = var_57_cast);
+            tensor<fp16, [512, 512]> var_98_to_fp16 = const()[name = tensor<string, []>("op_98_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3359232)))];
+            tensor<fp16, [512]> var_99_to_fp16 = const()[name = tensor<string, []>("op_99_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3883584)))];
+            tensor<fp16, [1, 1500, 512]> q_1_cast = linear(bias = var_99_to_fp16, weight = var_98_to_fp16, x = var_87_cast);
+            tensor<fp16, [512, 512]> var_102_to_fp16 = const()[name = tensor<string, []>("op_102_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3884672)))];
+            tensor<fp16, [512]> k_1_bias_0_to_fp16 = const()[name = tensor<string, []>("k_1_bias_0_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4409024)))];
+            tensor<fp16, [1, 1500, 512]> k_1_cast = linear(bias = k_1_bias_0_to_fp16, weight = var_102_to_fp16, x = var_87_cast);
+            tensor<fp16, [512, 512]> var_106_to_fp16 = const()[name = tensor<string, []>("op_106_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4410112)))];
+            tensor<fp16, [512]> var_107_to_fp16 = const()[name = tensor<string, []>("op_107_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4934464)))];
+            tensor<fp16, [1, 1500, 512]> v_1_cast = linear(bias = var_107_to_fp16, weight = var_106_to_fp16, x = var_87_cast);
+            tensor<int32, [4]> var_115 = const()[name = tensor<string, []>("op_115"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_116_cast = reshape(shape = var_115, x = q_1_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_42_to_fp16 = const()[name = tensor<string, []>("const_42_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> q_3_cast = mul(x = var_116_cast, y = const_42_to_fp16);
+            tensor<int32, [4]> var_122 = const()[name = tensor<string, []>("op_122"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_123_cast = reshape(shape = var_122, x = k_1_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_43_to_fp16 = const()[name = tensor<string, []>("const_43_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> k_3_cast = mul(x = var_123_cast, y = const_43_to_fp16);
+            tensor<int32, [4]> var_129 = const()[name = tensor<string, []>("op_129"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_130_cast = reshape(shape = var_129, x = v_1_cast);
+            tensor<int32, [4]> var_131 = const()[name = tensor<string, []>("op_131"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<bool, []> qk_1_transpose_x_0 = const()[name = tensor<string, []>("qk_1_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> qk_1_transpose_y_0 = const()[name = tensor<string, []>("qk_1_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<int32, [4]> transpose_12_perm_0 = const()[name = tensor<string, []>("transpose_12_perm_0"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [4]> transpose_13_perm_0 = const()[name = tensor<string, []>("transpose_13_perm_0"), val = tensor<int32, [4]>([0, 2, 3, 1])];
+            tensor<fp16, [1, 8, 64, 1500]> transpose_45 = transpose(perm = transpose_13_perm_0, x = k_3_cast);
+            tensor<fp16, [1, 8, 1500, 64]> transpose_46 = transpose(perm = transpose_12_perm_0, x = q_3_cast);
+            tensor<fp16, [1, 8, 1500, 1500]> qk_1_cast = matmul(transpose_x = qk_1_transpose_x_0, transpose_y = qk_1_transpose_y_0, x = transpose_46, y = transpose_45);
+            tensor<fp16, [1, 8, 1500, 1500]> var_135_cast = softmax(axis = var_70, x = qk_1_cast);
+            tensor<bool, []> var_137_transpose_x_0 = const()[name = tensor<string, []>("op_137_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> var_137_transpose_y_0 = const()[name = tensor<string, []>("op_137_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<fp16, [1, 8, 1500, 64]> transpose_47 = transpose(perm = var_131, x = var_130_cast);
+            tensor<fp16, [1, 8, 1500, 64]> var_137_cast = matmul(transpose_x = var_137_transpose_x_0, transpose_y = var_137_transpose_y_0, x = var_135_cast, y = transpose_47);
+            tensor<int32, [4]> var_138 = const()[name = tensor<string, []>("op_138"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [3]> concat_0 = const()[name = tensor<string, []>("concat_0"), val = tensor<int32, [3]>([1, 1500, 512])];
+            tensor<fp16, [1, 1500, 8, 64]> transpose_44 = transpose(perm = var_138, x = var_137_cast);
+            tensor<fp16, [1, 1500, 512]> x_11_cast = reshape(shape = concat_0, x = transpose_44);
+            tensor<fp16, [512, 512]> var_143_to_fp16 = const()[name = tensor<string, []>("op_143_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4935552)))];
+            tensor<fp16, [512]> var_144_to_fp16 = const()[name = tensor<string, []>("op_144_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(5459904)))];
+            tensor<fp16, [1, 1500, 512]> var_145_cast = linear(bias = var_144_to_fp16, weight = var_143_to_fp16, x = x_11_cast);
+            tensor<fp16, [1, 1500, 512]> x_13_cast = add(x = var_57_cast, y = var_145_cast);
+            tensor<int32, [1]> var_151_axes_0 = const()[name = tensor<string, []>("op_151_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_0_mlp_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_0_mlp_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(5460992)))];
+            tensor<fp16, [512]> blocks_0_mlp_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_0_mlp_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(5462080)))];
+            tensor<fp16, [1, 1500, 512]> var_151_cast = layer_norm(axes = var_151_axes_0, beta = blocks_0_mlp_ln_bias_to_fp16, epsilon = var_76_to_fp16, gamma = blocks_0_mlp_ln_weight_to_fp16, x = x_13_cast);
+            tensor<fp16, [2048, 512]> var_160_to_fp16 = const()[name = tensor<string, []>("op_160_to_fp16"), val = tensor<fp16, [2048, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(5463168)))];
+            tensor<fp16, [2048]> var_161_to_fp16 = const()[name = tensor<string, []>("op_161_to_fp16"), val = tensor<fp16, [2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(7560384)))];
+            tensor<fp16, [1, 1500, 2048]> input_9_cast = linear(bias = var_161_to_fp16, weight = var_160_to_fp16, x = var_151_cast);
+            tensor<string, []> x_17_mode_0 = const()[name = tensor<string, []>("x_17_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 1500, 2048]> x_17_cast = gelu(mode = x_17_mode_0, x = input_9_cast);
+            tensor<fp16, [512, 2048]> var_166_to_fp16 = const()[name = tensor<string, []>("op_166_to_fp16"), val = tensor<fp16, [512, 2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(7564544)))];
+            tensor<fp16, [512]> var_167_to_fp16 = const()[name = tensor<string, []>("op_167_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(9661760)))];
+            tensor<fp16, [1, 1500, 512]> var_168_cast = linear(bias = var_167_to_fp16, weight = var_166_to_fp16, x = x_17_cast);
+            tensor<fp16, [1, 1500, 512]> x_19_cast = add(x = x_13_cast, y = var_168_cast);
+            tensor<int32, []> var_177 = const()[name = tensor<string, []>("op_177"), val = tensor<int32, []>(-1)];
+            tensor<int32, [1]> var_194_axes_0 = const()[name = tensor<string, []>("op_194_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_1_attn_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_1_attn_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(9662848)))];
+            tensor<fp16, [512]> blocks_1_attn_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_1_attn_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(9663936)))];
+            tensor<fp16, []> var_183_to_fp16 = const()[name = tensor<string, []>("op_183_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_194_cast = layer_norm(axes = var_194_axes_0, beta = blocks_1_attn_ln_bias_to_fp16, epsilon = var_183_to_fp16, gamma = blocks_1_attn_ln_weight_to_fp16, x = x_19_cast);
+            tensor<fp16, [512, 512]> var_205_to_fp16 = const()[name = tensor<string, []>("op_205_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(9665024)))];
+            tensor<fp16, [512]> var_206_to_fp16 = const()[name = tensor<string, []>("op_206_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(10189376)))];
+            tensor<fp16, [1, 1500, 512]> q_5_cast = linear(bias = var_206_to_fp16, weight = var_205_to_fp16, x = var_194_cast);
+            tensor<fp16, [512, 512]> var_209_to_fp16 = const()[name = tensor<string, []>("op_209_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(10190464)))];
+            tensor<fp16, [512]> k_5_bias_0_to_fp16 = const()[name = tensor<string, []>("k_5_bias_0_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(10714816)))];
+            tensor<fp16, [1, 1500, 512]> k_5_cast = linear(bias = k_5_bias_0_to_fp16, weight = var_209_to_fp16, x = var_194_cast);
+            tensor<fp16, [512, 512]> var_213_to_fp16 = const()[name = tensor<string, []>("op_213_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(10715904)))];
+            tensor<fp16, [512]> var_214_to_fp16 = const()[name = tensor<string, []>("op_214_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(11240256)))];
+            tensor<fp16, [1, 1500, 512]> v_5_cast = linear(bias = var_214_to_fp16, weight = var_213_to_fp16, x = var_194_cast);
+            tensor<int32, [4]> var_222 = const()[name = tensor<string, []>("op_222"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_223_cast = reshape(shape = var_222, x = q_5_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_44_to_fp16 = const()[name = tensor<string, []>("const_44_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> q_7_cast = mul(x = var_223_cast, y = const_44_to_fp16);
+            tensor<int32, [4]> var_229 = const()[name = tensor<string, []>("op_229"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_230_cast = reshape(shape = var_229, x = k_5_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_45_to_fp16 = const()[name = tensor<string, []>("const_45_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> k_7_cast = mul(x = var_230_cast, y = const_45_to_fp16);
+            tensor<int32, [4]> var_236 = const()[name = tensor<string, []>("op_236"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_237_cast = reshape(shape = var_236, x = v_5_cast);
+            tensor<int32, [4]> var_238 = const()[name = tensor<string, []>("op_238"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<bool, []> qk_3_transpose_x_0 = const()[name = tensor<string, []>("qk_3_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> qk_3_transpose_y_0 = const()[name = tensor<string, []>("qk_3_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<int32, [4]> transpose_14_perm_0 = const()[name = tensor<string, []>("transpose_14_perm_0"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [4]> transpose_15_perm_0 = const()[name = tensor<string, []>("transpose_15_perm_0"), val = tensor<int32, [4]>([0, 2, 3, 1])];
+            tensor<fp16, [1, 8, 64, 1500]> transpose_41 = transpose(perm = transpose_15_perm_0, x = k_7_cast);
+            tensor<fp16, [1, 8, 1500, 64]> transpose_42 = transpose(perm = transpose_14_perm_0, x = q_7_cast);
+            tensor<fp16, [1, 8, 1500, 1500]> qk_3_cast = matmul(transpose_x = qk_3_transpose_x_0, transpose_y = qk_3_transpose_y_0, x = transpose_42, y = transpose_41);
+            tensor<fp16, [1, 8, 1500, 1500]> var_242_cast = softmax(axis = var_177, x = qk_3_cast);
+            tensor<bool, []> var_244_transpose_x_0 = const()[name = tensor<string, []>("op_244_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> var_244_transpose_y_0 = const()[name = tensor<string, []>("op_244_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<fp16, [1, 8, 1500, 64]> transpose_43 = transpose(perm = var_238, x = var_237_cast);
+            tensor<fp16, [1, 8, 1500, 64]> var_244_cast = matmul(transpose_x = var_244_transpose_x_0, transpose_y = var_244_transpose_y_0, x = var_242_cast, y = transpose_43);
+            tensor<int32, [4]> var_245 = const()[name = tensor<string, []>("op_245"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [3]> concat_1 = const()[name = tensor<string, []>("concat_1"), val = tensor<int32, [3]>([1, 1500, 512])];
+            tensor<fp16, [1, 1500, 8, 64]> transpose_40 = transpose(perm = var_245, x = var_244_cast);
+            tensor<fp16, [1, 1500, 512]> x_23_cast = reshape(shape = concat_1, x = transpose_40);
+            tensor<fp16, [512, 512]> var_250_to_fp16 = const()[name = tensor<string, []>("op_250_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(11241344)))];
+            tensor<fp16, [512]> var_251_to_fp16 = const()[name = tensor<string, []>("op_251_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(11765696)))];
+            tensor<fp16, [1, 1500, 512]> var_252_cast = linear(bias = var_251_to_fp16, weight = var_250_to_fp16, x = x_23_cast);
+            tensor<fp16, [1, 1500, 512]> x_25_cast = add(x = x_19_cast, y = var_252_cast);
+            tensor<int32, [1]> var_258_axes_0 = const()[name = tensor<string, []>("op_258_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_1_mlp_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_1_mlp_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(11766784)))];
+            tensor<fp16, [512]> blocks_1_mlp_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_1_mlp_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(11767872)))];
+            tensor<fp16, [1, 1500, 512]> var_258_cast = layer_norm(axes = var_258_axes_0, beta = blocks_1_mlp_ln_bias_to_fp16, epsilon = var_183_to_fp16, gamma = blocks_1_mlp_ln_weight_to_fp16, x = x_25_cast);
+            tensor<fp16, [2048, 512]> var_267_to_fp16 = const()[name = tensor<string, []>("op_267_to_fp16"), val = tensor<fp16, [2048, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(11768960)))];
+            tensor<fp16, [2048]> var_268_to_fp16 = const()[name = tensor<string, []>("op_268_to_fp16"), val = tensor<fp16, [2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(13866176)))];
+            tensor<fp16, [1, 1500, 2048]> input_17_cast = linear(bias = var_268_to_fp16, weight = var_267_to_fp16, x = var_258_cast);
+            tensor<string, []> x_29_mode_0 = const()[name = tensor<string, []>("x_29_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 1500, 2048]> x_29_cast = gelu(mode = x_29_mode_0, x = input_17_cast);
+            tensor<fp16, [512, 2048]> var_273_to_fp16 = const()[name = tensor<string, []>("op_273_to_fp16"), val = tensor<fp16, [512, 2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(13870336)))];
+            tensor<fp16, [512]> var_274_to_fp16 = const()[name = tensor<string, []>("op_274_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(15967552)))];
+            tensor<fp16, [1, 1500, 512]> var_275_cast = linear(bias = var_274_to_fp16, weight = var_273_to_fp16, x = x_29_cast);
+            tensor<fp16, [1, 1500, 512]> x_31_cast = add(x = x_25_cast, y = var_275_cast);
+            tensor<int32, []> var_284 = const()[name = tensor<string, []>("op_284"), val = tensor<int32, []>(-1)];
+            tensor<int32, [1]> var_301_axes_0 = const()[name = tensor<string, []>("op_301_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_2_attn_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_2_attn_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(15968640)))];
+            tensor<fp16, [512]> blocks_2_attn_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_2_attn_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(15969728)))];
+            tensor<fp16, []> var_290_to_fp16 = const()[name = tensor<string, []>("op_290_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_301_cast = layer_norm(axes = var_301_axes_0, beta = blocks_2_attn_ln_bias_to_fp16, epsilon = var_290_to_fp16, gamma = blocks_2_attn_ln_weight_to_fp16, x = x_31_cast);
+            tensor<fp16, [512, 512]> var_312_to_fp16 = const()[name = tensor<string, []>("op_312_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(15970816)))];
+            tensor<fp16, [512]> var_313_to_fp16 = const()[name = tensor<string, []>("op_313_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(16495168)))];
+            tensor<fp16, [1, 1500, 512]> q_9_cast = linear(bias = var_313_to_fp16, weight = var_312_to_fp16, x = var_301_cast);
+            tensor<fp16, [512, 512]> var_316_to_fp16 = const()[name = tensor<string, []>("op_316_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(16496256)))];
+            tensor<fp16, [512]> k_9_bias_0_to_fp16 = const()[name = tensor<string, []>("k_9_bias_0_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(17020608)))];
+            tensor<fp16, [1, 1500, 512]> k_9_cast = linear(bias = k_9_bias_0_to_fp16, weight = var_316_to_fp16, x = var_301_cast);
+            tensor<fp16, [512, 512]> var_320_to_fp16 = const()[name = tensor<string, []>("op_320_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(17021696)))];
+            tensor<fp16, [512]> var_321_to_fp16 = const()[name = tensor<string, []>("op_321_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(17546048)))];
+            tensor<fp16, [1, 1500, 512]> v_9_cast = linear(bias = var_321_to_fp16, weight = var_320_to_fp16, x = var_301_cast);
+            tensor<int32, [4]> var_329 = const()[name = tensor<string, []>("op_329"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_330_cast = reshape(shape = var_329, x = q_9_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_46_to_fp16 = const()[name = tensor<string, []>("const_46_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> q_11_cast = mul(x = var_330_cast, y = const_46_to_fp16);
+            tensor<int32, [4]> var_336 = const()[name = tensor<string, []>("op_336"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_337_cast = reshape(shape = var_336, x = k_9_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_47_to_fp16 = const()[name = tensor<string, []>("const_47_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> k_11_cast = mul(x = var_337_cast, y = const_47_to_fp16);
+            tensor<int32, [4]> var_343 = const()[name = tensor<string, []>("op_343"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_344_cast = reshape(shape = var_343, x = v_9_cast);
+            tensor<int32, [4]> var_345 = const()[name = tensor<string, []>("op_345"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<bool, []> qk_5_transpose_x_0 = const()[name = tensor<string, []>("qk_5_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> qk_5_transpose_y_0 = const()[name = tensor<string, []>("qk_5_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<int32, [4]> transpose_16_perm_0 = const()[name = tensor<string, []>("transpose_16_perm_0"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [4]> transpose_17_perm_0 = const()[name = tensor<string, []>("transpose_17_perm_0"), val = tensor<int32, [4]>([0, 2, 3, 1])];
+            tensor<fp16, [1, 8, 64, 1500]> transpose_37 = transpose(perm = transpose_17_perm_0, x = k_11_cast);
+            tensor<fp16, [1, 8, 1500, 64]> transpose_38 = transpose(perm = transpose_16_perm_0, x = q_11_cast);
+            tensor<fp16, [1, 8, 1500, 1500]> qk_5_cast = matmul(transpose_x = qk_5_transpose_x_0, transpose_y = qk_5_transpose_y_0, x = transpose_38, y = transpose_37);
+            tensor<fp16, [1, 8, 1500, 1500]> var_349_cast = softmax(axis = var_284, x = qk_5_cast);
+            tensor<bool, []> var_351_transpose_x_0 = const()[name = tensor<string, []>("op_351_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> var_351_transpose_y_0 = const()[name = tensor<string, []>("op_351_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<fp16, [1, 8, 1500, 64]> transpose_39 = transpose(perm = var_345, x = var_344_cast);
+            tensor<fp16, [1, 8, 1500, 64]> var_351_cast = matmul(transpose_x = var_351_transpose_x_0, transpose_y = var_351_transpose_y_0, x = var_349_cast, y = transpose_39);
+            tensor<int32, [4]> var_352 = const()[name = tensor<string, []>("op_352"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [3]> concat_2 = const()[name = tensor<string, []>("concat_2"), val = tensor<int32, [3]>([1, 1500, 512])];
+            tensor<fp16, [1, 1500, 8, 64]> transpose_36 = transpose(perm = var_352, x = var_351_cast);
+            tensor<fp16, [1, 1500, 512]> x_35_cast = reshape(shape = concat_2, x = transpose_36);
+            tensor<fp16, [512, 512]> var_357_to_fp16 = const()[name = tensor<string, []>("op_357_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(17547136)))];
+            tensor<fp16, [512]> var_358_to_fp16 = const()[name = tensor<string, []>("op_358_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(18071488)))];
+            tensor<fp16, [1, 1500, 512]> var_359_cast = linear(bias = var_358_to_fp16, weight = var_357_to_fp16, x = x_35_cast);
+            tensor<fp16, [1, 1500, 512]> x_37_cast = add(x = x_31_cast, y = var_359_cast);
+            tensor<int32, [1]> var_365_axes_0 = const()[name = tensor<string, []>("op_365_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_2_mlp_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_2_mlp_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(18072576)))];
+            tensor<fp16, [512]> blocks_2_mlp_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_2_mlp_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(18073664)))];
+            tensor<fp16, [1, 1500, 512]> var_365_cast = layer_norm(axes = var_365_axes_0, beta = blocks_2_mlp_ln_bias_to_fp16, epsilon = var_290_to_fp16, gamma = blocks_2_mlp_ln_weight_to_fp16, x = x_37_cast);
+            tensor<fp16, [2048, 512]> var_374_to_fp16 = const()[name = tensor<string, []>("op_374_to_fp16"), val = tensor<fp16, [2048, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(18074752)))];
+            tensor<fp16, [2048]> var_375_to_fp16 = const()[name = tensor<string, []>("op_375_to_fp16"), val = tensor<fp16, [2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(20171968)))];
+            tensor<fp16, [1, 1500, 2048]> input_25_cast = linear(bias = var_375_to_fp16, weight = var_374_to_fp16, x = var_365_cast);
+            tensor<string, []> x_41_mode_0 = const()[name = tensor<string, []>("x_41_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 1500, 2048]> x_41_cast = gelu(mode = x_41_mode_0, x = input_25_cast);
+            tensor<fp16, [512, 2048]> var_380_to_fp16 = const()[name = tensor<string, []>("op_380_to_fp16"), val = tensor<fp16, [512, 2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(20176128)))];
+            tensor<fp16, [512]> var_381_to_fp16 = const()[name = tensor<string, []>("op_381_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(22273344)))];
+            tensor<fp16, [1, 1500, 512]> var_382_cast = linear(bias = var_381_to_fp16, weight = var_380_to_fp16, x = x_41_cast);
+            tensor<fp16, [1, 1500, 512]> x_43_cast = add(x = x_37_cast, y = var_382_cast);
+            tensor<int32, []> var_391 = const()[name = tensor<string, []>("op_391"), val = tensor<int32, []>(-1)];
+            tensor<int32, [1]> var_408_axes_0 = const()[name = tensor<string, []>("op_408_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_3_attn_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_3_attn_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(22274432)))];
+            tensor<fp16, [512]> blocks_3_attn_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_3_attn_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(22275520)))];
+            tensor<fp16, []> var_397_to_fp16 = const()[name = tensor<string, []>("op_397_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_408_cast = layer_norm(axes = var_408_axes_0, beta = blocks_3_attn_ln_bias_to_fp16, epsilon = var_397_to_fp16, gamma = blocks_3_attn_ln_weight_to_fp16, x = x_43_cast);
+            tensor<fp16, [512, 512]> var_419_to_fp16 = const()[name = tensor<string, []>("op_419_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(22276608)))];
+            tensor<fp16, [512]> var_420_to_fp16 = const()[name = tensor<string, []>("op_420_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(22800960)))];
+            tensor<fp16, [1, 1500, 512]> q_13_cast = linear(bias = var_420_to_fp16, weight = var_419_to_fp16, x = var_408_cast);
+            tensor<fp16, [512, 512]> var_423_to_fp16 = const()[name = tensor<string, []>("op_423_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(22802048)))];
+            tensor<fp16, [512]> k_13_bias_0_to_fp16 = const()[name = tensor<string, []>("k_13_bias_0_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(23326400)))];
+            tensor<fp16, [1, 1500, 512]> k_13_cast = linear(bias = k_13_bias_0_to_fp16, weight = var_423_to_fp16, x = var_408_cast);
+            tensor<fp16, [512, 512]> var_427_to_fp16 = const()[name = tensor<string, []>("op_427_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(23327488)))];
+            tensor<fp16, [512]> var_428_to_fp16 = const()[name = tensor<string, []>("op_428_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(23851840)))];
+            tensor<fp16, [1, 1500, 512]> v_13_cast = linear(bias = var_428_to_fp16, weight = var_427_to_fp16, x = var_408_cast);
+            tensor<int32, [4]> var_436 = const()[name = tensor<string, []>("op_436"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_437_cast = reshape(shape = var_436, x = q_13_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_48_to_fp16 = const()[name = tensor<string, []>("const_48_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> q_15_cast = mul(x = var_437_cast, y = const_48_to_fp16);
+            tensor<int32, [4]> var_443 = const()[name = tensor<string, []>("op_443"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_444_cast = reshape(shape = var_443, x = k_13_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_49_to_fp16 = const()[name = tensor<string, []>("const_49_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> k_15_cast = mul(x = var_444_cast, y = const_49_to_fp16);
+            tensor<int32, [4]> var_450 = const()[name = tensor<string, []>("op_450"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_451_cast = reshape(shape = var_450, x = v_13_cast);
+            tensor<int32, [4]> var_452 = const()[name = tensor<string, []>("op_452"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<bool, []> qk_7_transpose_x_0 = const()[name = tensor<string, []>("qk_7_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> qk_7_transpose_y_0 = const()[name = tensor<string, []>("qk_7_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<int32, [4]> transpose_18_perm_0 = const()[name = tensor<string, []>("transpose_18_perm_0"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [4]> transpose_19_perm_0 = const()[name = tensor<string, []>("transpose_19_perm_0"), val = tensor<int32, [4]>([0, 2, 3, 1])];
+            tensor<fp16, [1, 8, 64, 1500]> transpose_33 = transpose(perm = transpose_19_perm_0, x = k_15_cast);
+            tensor<fp16, [1, 8, 1500, 64]> transpose_34 = transpose(perm = transpose_18_perm_0, x = q_15_cast);
+            tensor<fp16, [1, 8, 1500, 1500]> qk_7_cast = matmul(transpose_x = qk_7_transpose_x_0, transpose_y = qk_7_transpose_y_0, x = transpose_34, y = transpose_33);
+            tensor<fp16, [1, 8, 1500, 1500]> var_456_cast = softmax(axis = var_391, x = qk_7_cast);
+            tensor<bool, []> var_458_transpose_x_0 = const()[name = tensor<string, []>("op_458_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> var_458_transpose_y_0 = const()[name = tensor<string, []>("op_458_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<fp16, [1, 8, 1500, 64]> transpose_35 = transpose(perm = var_452, x = var_451_cast);
+            tensor<fp16, [1, 8, 1500, 64]> var_458_cast = matmul(transpose_x = var_458_transpose_x_0, transpose_y = var_458_transpose_y_0, x = var_456_cast, y = transpose_35);
+            tensor<int32, [4]> var_459 = const()[name = tensor<string, []>("op_459"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [3]> concat_3 = const()[name = tensor<string, []>("concat_3"), val = tensor<int32, [3]>([1, 1500, 512])];
+            tensor<fp16, [1, 1500, 8, 64]> transpose_32 = transpose(perm = var_459, x = var_458_cast);
+            tensor<fp16, [1, 1500, 512]> x_47_cast = reshape(shape = concat_3, x = transpose_32);
+            tensor<fp16, [512, 512]> var_464_to_fp16 = const()[name = tensor<string, []>("op_464_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(23852928)))];
+            tensor<fp16, [512]> var_465_to_fp16 = const()[name = tensor<string, []>("op_465_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(24377280)))];
+            tensor<fp16, [1, 1500, 512]> var_466_cast = linear(bias = var_465_to_fp16, weight = var_464_to_fp16, x = x_47_cast);
+            tensor<fp16, [1, 1500, 512]> x_49_cast = add(x = x_43_cast, y = var_466_cast);
+            tensor<int32, [1]> var_472_axes_0 = const()[name = tensor<string, []>("op_472_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_3_mlp_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_3_mlp_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(24378368)))];
+            tensor<fp16, [512]> blocks_3_mlp_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_3_mlp_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(24379456)))];
+            tensor<fp16, [1, 1500, 512]> var_472_cast = layer_norm(axes = var_472_axes_0, beta = blocks_3_mlp_ln_bias_to_fp16, epsilon = var_397_to_fp16, gamma = blocks_3_mlp_ln_weight_to_fp16, x = x_49_cast);
+            tensor<fp16, [2048, 512]> var_481_to_fp16 = const()[name = tensor<string, []>("op_481_to_fp16"), val = tensor<fp16, [2048, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(24380544)))];
+            tensor<fp16, [2048]> var_482_to_fp16 = const()[name = tensor<string, []>("op_482_to_fp16"), val = tensor<fp16, [2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(26477760)))];
+            tensor<fp16, [1, 1500, 2048]> input_33_cast = linear(bias = var_482_to_fp16, weight = var_481_to_fp16, x = var_472_cast);
+            tensor<string, []> x_53_mode_0 = const()[name = tensor<string, []>("x_53_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 1500, 2048]> x_53_cast = gelu(mode = x_53_mode_0, x = input_33_cast);
+            tensor<fp16, [512, 2048]> var_487_to_fp16 = const()[name = tensor<string, []>("op_487_to_fp16"), val = tensor<fp16, [512, 2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(26481920)))];
+            tensor<fp16, [512]> var_488_to_fp16 = const()[name = tensor<string, []>("op_488_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(28579136)))];
+            tensor<fp16, [1, 1500, 512]> var_489_cast = linear(bias = var_488_to_fp16, weight = var_487_to_fp16, x = x_53_cast);
+            tensor<fp16, [1, 1500, 512]> x_55_cast = add(x = x_49_cast, y = var_489_cast);
+            tensor<int32, []> var_498 = const()[name = tensor<string, []>("op_498"), val = tensor<int32, []>(-1)];
+            tensor<int32, [1]> var_515_axes_0 = const()[name = tensor<string, []>("op_515_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_4_attn_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_4_attn_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(28580224)))];
+            tensor<fp16, [512]> blocks_4_attn_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_4_attn_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(28581312)))];
+            tensor<fp16, []> var_504_to_fp16 = const()[name = tensor<string, []>("op_504_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_515_cast = layer_norm(axes = var_515_axes_0, beta = blocks_4_attn_ln_bias_to_fp16, epsilon = var_504_to_fp16, gamma = blocks_4_attn_ln_weight_to_fp16, x = x_55_cast);
+            tensor<fp16, [512, 512]> var_526_to_fp16 = const()[name = tensor<string, []>("op_526_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(28582400)))];
+            tensor<fp16, [512]> var_527_to_fp16 = const()[name = tensor<string, []>("op_527_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(29106752)))];
+            tensor<fp16, [1, 1500, 512]> q_17_cast = linear(bias = var_527_to_fp16, weight = var_526_to_fp16, x = var_515_cast);
+            tensor<fp16, [512, 512]> var_530_to_fp16 = const()[name = tensor<string, []>("op_530_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(29107840)))];
+            tensor<fp16, [512]> k_17_bias_0_to_fp16 = const()[name = tensor<string, []>("k_17_bias_0_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(29632192)))];
+            tensor<fp16, [1, 1500, 512]> k_17_cast = linear(bias = k_17_bias_0_to_fp16, weight = var_530_to_fp16, x = var_515_cast);
+            tensor<fp16, [512, 512]> var_534_to_fp16 = const()[name = tensor<string, []>("op_534_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(29633280)))];
+            tensor<fp16, [512]> var_535_to_fp16 = const()[name = tensor<string, []>("op_535_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(30157632)))];
+            tensor<fp16, [1, 1500, 512]> v_17_cast = linear(bias = var_535_to_fp16, weight = var_534_to_fp16, x = var_515_cast);
+            tensor<int32, [4]> var_543 = const()[name = tensor<string, []>("op_543"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_544_cast = reshape(shape = var_543, x = q_17_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_50_to_fp16 = const()[name = tensor<string, []>("const_50_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> q_19_cast = mul(x = var_544_cast, y = const_50_to_fp16);
+            tensor<int32, [4]> var_550 = const()[name = tensor<string, []>("op_550"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_551_cast = reshape(shape = var_550, x = k_17_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_51_to_fp16 = const()[name = tensor<string, []>("const_51_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> k_19_cast = mul(x = var_551_cast, y = const_51_to_fp16);
+            tensor<int32, [4]> var_557 = const()[name = tensor<string, []>("op_557"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_558_cast = reshape(shape = var_557, x = v_17_cast);
+            tensor<int32, [4]> var_559 = const()[name = tensor<string, []>("op_559"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<bool, []> qk_9_transpose_x_0 = const()[name = tensor<string, []>("qk_9_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> qk_9_transpose_y_0 = const()[name = tensor<string, []>("qk_9_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<int32, [4]> transpose_20_perm_0 = const()[name = tensor<string, []>("transpose_20_perm_0"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [4]> transpose_21_perm_0 = const()[name = tensor<string, []>("transpose_21_perm_0"), val = tensor<int32, [4]>([0, 2, 3, 1])];
+            tensor<fp16, [1, 8, 64, 1500]> transpose_29 = transpose(perm = transpose_21_perm_0, x = k_19_cast);
+            tensor<fp16, [1, 8, 1500, 64]> transpose_30 = transpose(perm = transpose_20_perm_0, x = q_19_cast);
+            tensor<fp16, [1, 8, 1500, 1500]> qk_9_cast = matmul(transpose_x = qk_9_transpose_x_0, transpose_y = qk_9_transpose_y_0, x = transpose_30, y = transpose_29);
+            tensor<fp16, [1, 8, 1500, 1500]> var_563_cast = softmax(axis = var_498, x = qk_9_cast);
+            tensor<bool, []> var_565_transpose_x_0 = const()[name = tensor<string, []>("op_565_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> var_565_transpose_y_0 = const()[name = tensor<string, []>("op_565_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<fp16, [1, 8, 1500, 64]> transpose_31 = transpose(perm = var_559, x = var_558_cast);
+            tensor<fp16, [1, 8, 1500, 64]> var_565_cast = matmul(transpose_x = var_565_transpose_x_0, transpose_y = var_565_transpose_y_0, x = var_563_cast, y = transpose_31);
+            tensor<int32, [4]> var_566 = const()[name = tensor<string, []>("op_566"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [3]> concat_4 = const()[name = tensor<string, []>("concat_4"), val = tensor<int32, [3]>([1, 1500, 512])];
+            tensor<fp16, [1, 1500, 8, 64]> transpose_28 = transpose(perm = var_566, x = var_565_cast);
+            tensor<fp16, [1, 1500, 512]> x_59_cast = reshape(shape = concat_4, x = transpose_28);
+            tensor<fp16, [512, 512]> var_571_to_fp16 = const()[name = tensor<string, []>("op_571_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(30158720)))];
+            tensor<fp16, [512]> var_572_to_fp16 = const()[name = tensor<string, []>("op_572_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(30683072)))];
+            tensor<fp16, [1, 1500, 512]> var_573_cast = linear(bias = var_572_to_fp16, weight = var_571_to_fp16, x = x_59_cast);
+            tensor<fp16, [1, 1500, 512]> x_61_cast = add(x = x_55_cast, y = var_573_cast);
+            tensor<int32, [1]> var_579_axes_0 = const()[name = tensor<string, []>("op_579_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_4_mlp_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_4_mlp_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(30684160)))];
+            tensor<fp16, [512]> blocks_4_mlp_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_4_mlp_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(30685248)))];
+            tensor<fp16, [1, 1500, 512]> var_579_cast = layer_norm(axes = var_579_axes_0, beta = blocks_4_mlp_ln_bias_to_fp16, epsilon = var_504_to_fp16, gamma = blocks_4_mlp_ln_weight_to_fp16, x = x_61_cast);
+            tensor<fp16, [2048, 512]> var_588_to_fp16 = const()[name = tensor<string, []>("op_588_to_fp16"), val = tensor<fp16, [2048, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(30686336)))];
+            tensor<fp16, [2048]> var_589_to_fp16 = const()[name = tensor<string, []>("op_589_to_fp16"), val = tensor<fp16, [2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(32783552)))];
+            tensor<fp16, [1, 1500, 2048]> input_41_cast = linear(bias = var_589_to_fp16, weight = var_588_to_fp16, x = var_579_cast);
+            tensor<string, []> x_65_mode_0 = const()[name = tensor<string, []>("x_65_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 1500, 2048]> x_65_cast = gelu(mode = x_65_mode_0, x = input_41_cast);
+            tensor<fp16, [512, 2048]> var_594_to_fp16 = const()[name = tensor<string, []>("op_594_to_fp16"), val = tensor<fp16, [512, 2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(32787712)))];
+            tensor<fp16, [512]> var_595_to_fp16 = const()[name = tensor<string, []>("op_595_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(34884928)))];
+            tensor<fp16, [1, 1500, 512]> var_596_cast = linear(bias = var_595_to_fp16, weight = var_594_to_fp16, x = x_65_cast);
+            tensor<fp16, [1, 1500, 512]> x_67_cast = add(x = x_61_cast, y = var_596_cast);
+            tensor<int32, []> var_605 = const()[name = tensor<string, []>("op_605"), val = tensor<int32, []>(-1)];
+            tensor<int32, [1]> var_622_axes_0 = const()[name = tensor<string, []>("op_622_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_5_attn_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_5_attn_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(34886016)))];
+            tensor<fp16, [512]> blocks_5_attn_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_5_attn_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(34887104)))];
+            tensor<fp16, []> var_611_to_fp16 = const()[name = tensor<string, []>("op_611_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_622_cast = layer_norm(axes = var_622_axes_0, beta = blocks_5_attn_ln_bias_to_fp16, epsilon = var_611_to_fp16, gamma = blocks_5_attn_ln_weight_to_fp16, x = x_67_cast);
+            tensor<fp16, [512, 512]> var_633_to_fp16 = const()[name = tensor<string, []>("op_633_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(34888192)))];
+            tensor<fp16, [512]> var_634_to_fp16 = const()[name = tensor<string, []>("op_634_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(35412544)))];
+            tensor<fp16, [1, 1500, 512]> q_21_cast = linear(bias = var_634_to_fp16, weight = var_633_to_fp16, x = var_622_cast);
+            tensor<fp16, [512, 512]> var_637_to_fp16 = const()[name = tensor<string, []>("op_637_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(35413632)))];
+            tensor<fp16, [512]> k_21_bias_0_to_fp16 = const()[name = tensor<string, []>("k_21_bias_0_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(35937984)))];
+            tensor<fp16, [1, 1500, 512]> k_21_cast = linear(bias = k_21_bias_0_to_fp16, weight = var_637_to_fp16, x = var_622_cast);
+            tensor<fp16, [512, 512]> var_641_to_fp16 = const()[name = tensor<string, []>("op_641_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(35939072)))];
+            tensor<fp16, [512]> var_642_to_fp16 = const()[name = tensor<string, []>("op_642_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(36463424)))];
+            tensor<fp16, [1, 1500, 512]> v_21_cast = linear(bias = var_642_to_fp16, weight = var_641_to_fp16, x = var_622_cast);
+            tensor<int32, [4]> var_650 = const()[name = tensor<string, []>("op_650"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_651_cast = reshape(shape = var_650, x = q_21_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_52_to_fp16 = const()[name = tensor<string, []>("const_52_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> q_cast = mul(x = var_651_cast, y = const_52_to_fp16);
+            tensor<int32, [4]> var_657 = const()[name = tensor<string, []>("op_657"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_658_cast = reshape(shape = var_657, x = k_21_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_53_to_fp16 = const()[name = tensor<string, []>("const_53_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> k_cast = mul(x = var_658_cast, y = const_53_to_fp16);
+            tensor<int32, [4]> var_664 = const()[name = tensor<string, []>("op_664"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_665_cast = reshape(shape = var_664, x = v_21_cast);
+            tensor<int32, [4]> var_666 = const()[name = tensor<string, []>("op_666"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<bool, []> qk_transpose_x_0 = const()[name = tensor<string, []>("qk_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> qk_transpose_y_0 = const()[name = tensor<string, []>("qk_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<int32, [4]> transpose_22_perm_0 = const()[name = tensor<string, []>("transpose_22_perm_0"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [4]> transpose_23_perm_0 = const()[name = tensor<string, []>("transpose_23_perm_0"), val = tensor<int32, [4]>([0, 2, 3, 1])];
+            tensor<fp16, [1, 8, 64, 1500]> transpose_25 = transpose(perm = transpose_23_perm_0, x = k_cast);
+            tensor<fp16, [1, 8, 1500, 64]> transpose_26 = transpose(perm = transpose_22_perm_0, x = q_cast);
+            tensor<fp16, [1, 8, 1500, 1500]> qk_cast = matmul(transpose_x = qk_transpose_x_0, transpose_y = qk_transpose_y_0, x = transpose_26, y = transpose_25);
+            tensor<fp16, [1, 8, 1500, 1500]> var_670_cast = softmax(axis = var_605, x = qk_cast);
+            tensor<bool, []> var_672_transpose_x_0 = const()[name = tensor<string, []>("op_672_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> var_672_transpose_y_0 = const()[name = tensor<string, []>("op_672_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<fp16, [1, 8, 1500, 64]> transpose_27 = transpose(perm = var_666, x = var_665_cast);
+            tensor<fp16, [1, 8, 1500, 64]> var_672_cast = matmul(transpose_x = var_672_transpose_x_0, transpose_y = var_672_transpose_y_0, x = var_670_cast, y = transpose_27);
+            tensor<int32, [4]> var_673 = const()[name = tensor<string, []>("op_673"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [3]> concat_5 = const()[name = tensor<string, []>("concat_5"), val = tensor<int32, [3]>([1, 1500, 512])];
+            tensor<fp16, [1, 1500, 8, 64]> transpose_24 = transpose(perm = var_673, x = var_672_cast);
+            tensor<fp16, [1, 1500, 512]> x_71_cast = reshape(shape = concat_5, x = transpose_24);
+            tensor<fp16, [512, 512]> var_678_to_fp16 = const()[name = tensor<string, []>("op_678_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(36464512)))];
+            tensor<fp16, [512]> var_679_to_fp16 = const()[name = tensor<string, []>("op_679_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(36988864)))];
+            tensor<fp16, [1, 1500, 512]> var_680_cast = linear(bias = var_679_to_fp16, weight = var_678_to_fp16, x = x_71_cast);
+            tensor<fp16, [1, 1500, 512]> x_73_cast = add(x = x_67_cast, y = var_680_cast);
+            tensor<int32, [1]> var_686_axes_0 = const()[name = tensor<string, []>("op_686_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_5_mlp_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_5_mlp_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(36989952)))];
+            tensor<fp16, [512]> blocks_5_mlp_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_5_mlp_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(36991040)))];
+            tensor<fp16, [1, 1500, 512]> var_686_cast = layer_norm(axes = var_686_axes_0, beta = blocks_5_mlp_ln_bias_to_fp16, epsilon = var_611_to_fp16, gamma = blocks_5_mlp_ln_weight_to_fp16, x = x_73_cast);
+            tensor<fp16, [2048, 512]> var_695_to_fp16 = const()[name = tensor<string, []>("op_695_to_fp16"), val = tensor<fp16, [2048, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(36992128)))];
+            tensor<fp16, [2048]> var_696_to_fp16 = const()[name = tensor<string, []>("op_696_to_fp16"), val = tensor<fp16, [2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(39089344)))];
+            tensor<fp16, [1, 1500, 2048]> input_49_cast = linear(bias = var_696_to_fp16, weight = var_695_to_fp16, x = var_686_cast);
+            tensor<string, []> x_77_mode_0 = const()[name = tensor<string, []>("x_77_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 1500, 2048]> x_77_cast = gelu(mode = x_77_mode_0, x = input_49_cast);
+            tensor<fp16, [512, 2048]> var_701_to_fp16 = const()[name = tensor<string, []>("op_701_to_fp16"), val = tensor<fp16, [512, 2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(39093504)))];
+            tensor<fp16, [512]> var_702_to_fp16 = const()[name = tensor<string, []>("op_702_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(41190720)))];
+            tensor<fp16, [1, 1500, 512]> var_703_cast = linear(bias = var_702_to_fp16, weight = var_701_to_fp16, x = x_77_cast);
+            tensor<fp16, [1, 1500, 512]> x_cast = add(x = x_73_cast, y = var_703_cast);
+            tensor<int32, [1]> var_716_axes_0 = const()[name = tensor<string, []>("op_716_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> ln_post_weight_to_fp16 = const()[name = tensor<string, []>("ln_post_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(41191808)))];
+            tensor<fp16, [512]> ln_post_bias_to_fp16 = const()[name = tensor<string, []>("ln_post_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(41192896)))];
+            tensor<fp16, []> var_707_to_fp16 = const()[name = tensor<string, []>("op_707_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_716_cast = layer_norm(axes = var_716_axes_0, beta = ln_post_bias_to_fp16, epsilon = var_707_to_fp16, gamma = ln_post_weight_to_fp16, x = x_cast);
+            tensor<string, []> var_716_cast_to_fp32_dtype_0 = const()[name = tensor<string, []>("op_716_cast_to_fp32_dtype_0"), val = tensor<string, []>("fp32")];
+            tensor<fp32, [1, 1500, 512]> output = cast(dtype = var_716_cast_to_fp32_dtype_0, x = var_716_cast);
+        } -> (output);
+}

ggml-base-encoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa38c9d90f01effad1990d1f17e67cd8382c6486b8e5a4dbacf44439cd838a38
+size 41193984

ggml-base.en-encoder.mlmodelc.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8cf860309e2449e2bdc8be834cf838ab2565747ecc8c0ef914ef5975115e192b
+oid sha256:0634e109c4ec900dc6ffc3305ac8367822ade13fcc10425ec0db0dbfcebb12aa
 size 37950917

ggml-base.en-encoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2441ae34fc7d12946dba7b63379063e856ffc7c3e11ba5f7533efb1450562ca6
+size 207

ggml-base.en-encoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05fe28591b40616fa0c34ad7b853133623f5300923ec812acb11459c411acf3b
+size 149

ggml-base.en-encoder.mlmodelc/metadata.json

@@ -0,0 +1,64 @@
+[
+  {
+    "metadataOutputVersion" : "3.0",
+    "storagePrecision" : "Float16",
+    "outputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32)",
+        "shortDescription" : "",
+        "shape" : "[]",
+        "name" : "output",
+        "type" : "MultiArray"
+      }
+    ],
+    "modelParameters" : [
+
+    ],
+    "specificationVersion" : 6,
+    "mlProgramOperationTypeHistogram" : {
+      "Linear" : 36,
+      "Matmul" : 12,
+      "Cast" : 2,
+      "Conv" : 2,
+      "Softmax" : 6,
+      "Add" : 13,
+      "LayerNorm" : 13,
+      "Mul" : 12,
+      "Transpose" : 25,
+      "Gelu" : 8,
+      "Reshape" : 24
+    },
+    "computePrecision" : "Mixed (Float16, Float32, Int32)",
+    "isUpdatable" : "0",
+    "availability" : {
+      "macOS" : "12.0",
+      "tvOS" : "15.0",
+      "watchOS" : "8.0",
+      "iOS" : "15.0",
+      "macCatalyst" : "15.0"
+    },
+    "modelType" : {
+      "name" : "MLModelType_mlProgram"
+    },
+    "userDefinedMetadata" : {
+
+    },
+    "inputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32 1 × 80 × 3000)",
+        "shortDescription" : "",
+        "shape" : "[1, 80, 3000]",
+        "name" : "logmel_data",
+        "type" : "MultiArray"
+      }
+    ],
+    "generatedClassName" : "coreml_encoder_base_en",
+    "method" : "predict"
+  }
+]

ggml-base.en-encoder.mlmodelc/model.mil

@@ -0,0 +1,393 @@
+program(1.0)
+[buildInfo = dict<tensor<string, []>, tensor<string, []>>({{"coremlc-component-MIL", "4.28.4"}, {"coremlc-version", "1436.100.10"}})]
+{
+    func main<ios15>(tensor<fp32, [1, 80, 3000]> logmel_data) {
+            tensor<int32, []> var_20 = const()[name = tensor<string, []>("op_20"), val = tensor<int32, []>(1)];
+            tensor<int32, [1]> var_28 = const()[name = tensor<string, []>("op_28"), val = tensor<int32, [1]>([1])];
+            tensor<int32, [1]> var_30 = const()[name = tensor<string, []>("op_30"), val = tensor<int32, [1]>([1])];
+            tensor<string, []> var_32_pad_type_0 = const()[name = tensor<string, []>("op_32_pad_type_0"), val = tensor<string, []>("custom")];
+            tensor<int32, [2]> var_32_pad_0 = const()[name = tensor<string, []>("op_32_pad_0"), val = tensor<int32, [2]>([1, 1])];
+            tensor<string, []> logmel_data_to_fp16_dtype_0 = const()[name = tensor<string, []>("logmel_data_to_fp16_dtype_0"), val = tensor<string, []>("fp16")];
+            tensor<fp16, [512, 80, 3]> weight_3_to_fp16 = const()[name = tensor<string, []>("weight_3_to_fp16"), val = tensor<fp16, [512, 80, 3]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(64)))];
+            tensor<fp16, [512]> bias_3_to_fp16 = const()[name = tensor<string, []>("bias_3_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(245888)))];
+            tensor<fp16, [1, 80, 3000]> cast_187 = cast(dtype = logmel_data_to_fp16_dtype_0, x = logmel_data);
+            tensor<fp16, [1, 512, 3000]> var_32_cast = conv(bias = bias_3_to_fp16, dilations = var_30, groups = var_20, pad = var_32_pad_0, pad_type = var_32_pad_type_0, strides = var_28, weight = weight_3_to_fp16, x = cast_187);
+            tensor<string, []> input_1_mode_0 = const()[name = tensor<string, []>("input_1_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 512, 3000]> input_1_cast = gelu(mode = input_1_mode_0, x = var_32_cast);
+            tensor<int32, []> var_36 = const()[name = tensor<string, []>("op_36"), val = tensor<int32, []>(1)];
+            tensor<int32, [1]> var_45 = const()[name = tensor<string, []>("op_45"), val = tensor<int32, [1]>([2])];
+            tensor<int32, [1]> var_47 = const()[name = tensor<string, []>("op_47"), val = tensor<int32, [1]>([1])];
+            tensor<string, []> var_49_pad_type_0 = const()[name = tensor<string, []>("op_49_pad_type_0"), val = tensor<string, []>("custom")];
+            tensor<int32, [2]> var_49_pad_0 = const()[name = tensor<string, []>("op_49_pad_0"), val = tensor<int32, [2]>([1, 1])];
+            tensor<fp16, [512, 512, 3]> weight_7_to_fp16 = const()[name = tensor<string, []>("weight_7_to_fp16"), val = tensor<fp16, [512, 512, 3]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(246976)))];
+            tensor<fp16, [512]> bias_7_to_fp16 = const()[name = tensor<string, []>("bias_7_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1819904)))];
+            tensor<fp16, [1, 512, 1500]> var_49_cast = conv(bias = bias_7_to_fp16, dilations = var_47, groups = var_36, pad = var_49_pad_0, pad_type = var_49_pad_type_0, strides = var_45, weight = weight_7_to_fp16, x = input_1_cast);
+            tensor<string, []> x_3_mode_0 = const()[name = tensor<string, []>("x_3_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 512, 1500]> x_3_cast = gelu(mode = x_3_mode_0, x = var_49_cast);
+            tensor<int32, [3]> var_54 = const()[name = tensor<string, []>("op_54"), val = tensor<int32, [3]>([0, 2, 1])];
+            tensor<fp16, [1500, 512]> positional_embedding_to_fp16 = const()[name = tensor<string, []>("positional_embedding_to_fp16"), val = tensor<fp16, [1500, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1820992)))];
+            tensor<fp16, [1, 1500, 512]> transpose_48 = transpose(perm = var_54, x = x_3_cast);
+            tensor<fp16, [1, 1500, 512]> var_57_cast = add(x = transpose_48, y = positional_embedding_to_fp16);
+            tensor<int32, []> var_70 = const()[name = tensor<string, []>("op_70"), val = tensor<int32, []>(-1)];
+            tensor<int32, [1]> var_87_axes_0 = const()[name = tensor<string, []>("op_87_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_0_attn_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_0_attn_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3357056)))];
+            tensor<fp16, [512]> blocks_0_attn_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_0_attn_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3358144)))];
+            tensor<fp16, []> var_76_to_fp16 = const()[name = tensor<string, []>("op_76_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_87_cast = layer_norm(axes = var_87_axes_0, beta = blocks_0_attn_ln_bias_to_fp16, epsilon = var_76_to_fp16, gamma = blocks_0_attn_ln_weight_to_fp16, x = var_57_cast);
+            tensor<fp16, [512, 512]> var_98_to_fp16 = const()[name = tensor<string, []>("op_98_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3359232)))];
+            tensor<fp16, [512]> var_99_to_fp16 = const()[name = tensor<string, []>("op_99_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3883584)))];
+            tensor<fp16, [1, 1500, 512]> q_1_cast = linear(bias = var_99_to_fp16, weight = var_98_to_fp16, x = var_87_cast);
+            tensor<fp16, [512, 512]> var_102_to_fp16 = const()[name = tensor<string, []>("op_102_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3884672)))];
+            tensor<fp16, [512]> k_1_bias_0_to_fp16 = const()[name = tensor<string, []>("k_1_bias_0_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4409024)))];
+            tensor<fp16, [1, 1500, 512]> k_1_cast = linear(bias = k_1_bias_0_to_fp16, weight = var_102_to_fp16, x = var_87_cast);
+            tensor<fp16, [512, 512]> var_106_to_fp16 = const()[name = tensor<string, []>("op_106_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4410112)))];
+            tensor<fp16, [512]> var_107_to_fp16 = const()[name = tensor<string, []>("op_107_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4934464)))];
+            tensor<fp16, [1, 1500, 512]> v_1_cast = linear(bias = var_107_to_fp16, weight = var_106_to_fp16, x = var_87_cast);
+            tensor<int32, [4]> var_115 = const()[name = tensor<string, []>("op_115"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_116_cast = reshape(shape = var_115, x = q_1_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_42_to_fp16 = const()[name = tensor<string, []>("const_42_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> q_3_cast = mul(x = var_116_cast, y = const_42_to_fp16);
+            tensor<int32, [4]> var_122 = const()[name = tensor<string, []>("op_122"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_123_cast = reshape(shape = var_122, x = k_1_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_43_to_fp16 = const()[name = tensor<string, []>("const_43_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> k_3_cast = mul(x = var_123_cast, y = const_43_to_fp16);
+            tensor<int32, [4]> var_129 = const()[name = tensor<string, []>("op_129"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_130_cast = reshape(shape = var_129, x = v_1_cast);
+            tensor<int32, [4]> var_131 = const()[name = tensor<string, []>("op_131"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<bool, []> qk_1_transpose_x_0 = const()[name = tensor<string, []>("qk_1_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> qk_1_transpose_y_0 = const()[name = tensor<string, []>("qk_1_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<int32, [4]> transpose_12_perm_0 = const()[name = tensor<string, []>("transpose_12_perm_0"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [4]> transpose_13_perm_0 = const()[name = tensor<string, []>("transpose_13_perm_0"), val = tensor<int32, [4]>([0, 2, 3, 1])];
+            tensor<fp16, [1, 8, 64, 1500]> transpose_45 = transpose(perm = transpose_13_perm_0, x = k_3_cast);
+            tensor<fp16, [1, 8, 1500, 64]> transpose_46 = transpose(perm = transpose_12_perm_0, x = q_3_cast);
+            tensor<fp16, [1, 8, 1500, 1500]> qk_1_cast = matmul(transpose_x = qk_1_transpose_x_0, transpose_y = qk_1_transpose_y_0, x = transpose_46, y = transpose_45);
+            tensor<fp16, [1, 8, 1500, 1500]> var_135_cast = softmax(axis = var_70, x = qk_1_cast);
+            tensor<bool, []> var_137_transpose_x_0 = const()[name = tensor<string, []>("op_137_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> var_137_transpose_y_0 = const()[name = tensor<string, []>("op_137_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<fp16, [1, 8, 1500, 64]> transpose_47 = transpose(perm = var_131, x = var_130_cast);
+            tensor<fp16, [1, 8, 1500, 64]> var_137_cast = matmul(transpose_x = var_137_transpose_x_0, transpose_y = var_137_transpose_y_0, x = var_135_cast, y = transpose_47);
+            tensor<int32, [4]> var_138 = const()[name = tensor<string, []>("op_138"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [3]> concat_0 = const()[name = tensor<string, []>("concat_0"), val = tensor<int32, [3]>([1, 1500, 512])];
+            tensor<fp16, [1, 1500, 8, 64]> transpose_44 = transpose(perm = var_138, x = var_137_cast);
+            tensor<fp16, [1, 1500, 512]> x_11_cast = reshape(shape = concat_0, x = transpose_44);
+            tensor<fp16, [512, 512]> var_143_to_fp16 = const()[name = tensor<string, []>("op_143_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4935552)))];
+            tensor<fp16, [512]> var_144_to_fp16 = const()[name = tensor<string, []>("op_144_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(5459904)))];
+            tensor<fp16, [1, 1500, 512]> var_145_cast = linear(bias = var_144_to_fp16, weight = var_143_to_fp16, x = x_11_cast);
+            tensor<fp16, [1, 1500, 512]> x_13_cast = add(x = var_57_cast, y = var_145_cast);
+            tensor<int32, [1]> var_151_axes_0 = const()[name = tensor<string, []>("op_151_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_0_mlp_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_0_mlp_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(5460992)))];
+            tensor<fp16, [512]> blocks_0_mlp_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_0_mlp_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(5462080)))];
+            tensor<fp16, [1, 1500, 512]> var_151_cast = layer_norm(axes = var_151_axes_0, beta = blocks_0_mlp_ln_bias_to_fp16, epsilon = var_76_to_fp16, gamma = blocks_0_mlp_ln_weight_to_fp16, x = x_13_cast);
+            tensor<fp16, [2048, 512]> var_160_to_fp16 = const()[name = tensor<string, []>("op_160_to_fp16"), val = tensor<fp16, [2048, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(5463168)))];
+            tensor<fp16, [2048]> var_161_to_fp16 = const()[name = tensor<string, []>("op_161_to_fp16"), val = tensor<fp16, [2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(7560384)))];
+            tensor<fp16, [1, 1500, 2048]> input_9_cast = linear(bias = var_161_to_fp16, weight = var_160_to_fp16, x = var_151_cast);
+            tensor<string, []> x_17_mode_0 = const()[name = tensor<string, []>("x_17_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 1500, 2048]> x_17_cast = gelu(mode = x_17_mode_0, x = input_9_cast);
+            tensor<fp16, [512, 2048]> var_166_to_fp16 = const()[name = tensor<string, []>("op_166_to_fp16"), val = tensor<fp16, [512, 2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(7564544)))];
+            tensor<fp16, [512]> var_167_to_fp16 = const()[name = tensor<string, []>("op_167_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(9661760)))];
+            tensor<fp16, [1, 1500, 512]> var_168_cast = linear(bias = var_167_to_fp16, weight = var_166_to_fp16, x = x_17_cast);
+            tensor<fp16, [1, 1500, 512]> x_19_cast = add(x = x_13_cast, y = var_168_cast);
+            tensor<int32, []> var_177 = const()[name = tensor<string, []>("op_177"), val = tensor<int32, []>(-1)];
+            tensor<int32, [1]> var_194_axes_0 = const()[name = tensor<string, []>("op_194_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_1_attn_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_1_attn_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(9662848)))];
+            tensor<fp16, [512]> blocks_1_attn_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_1_attn_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(9663936)))];
+            tensor<fp16, []> var_183_to_fp16 = const()[name = tensor<string, []>("op_183_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_194_cast = layer_norm(axes = var_194_axes_0, beta = blocks_1_attn_ln_bias_to_fp16, epsilon = var_183_to_fp16, gamma = blocks_1_attn_ln_weight_to_fp16, x = x_19_cast);
+            tensor<fp16, [512, 512]> var_205_to_fp16 = const()[name = tensor<string, []>("op_205_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(9665024)))];
+            tensor<fp16, [512]> var_206_to_fp16 = const()[name = tensor<string, []>("op_206_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(10189376)))];
+            tensor<fp16, [1, 1500, 512]> q_5_cast = linear(bias = var_206_to_fp16, weight = var_205_to_fp16, x = var_194_cast);
+            tensor<fp16, [512, 512]> var_209_to_fp16 = const()[name = tensor<string, []>("op_209_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(10190464)))];
+            tensor<fp16, [512]> k_5_bias_0_to_fp16 = const()[name = tensor<string, []>("k_5_bias_0_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(10714816)))];
+            tensor<fp16, [1, 1500, 512]> k_5_cast = linear(bias = k_5_bias_0_to_fp16, weight = var_209_to_fp16, x = var_194_cast);
+            tensor<fp16, [512, 512]> var_213_to_fp16 = const()[name = tensor<string, []>("op_213_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(10715904)))];
+            tensor<fp16, [512]> var_214_to_fp16 = const()[name = tensor<string, []>("op_214_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(11240256)))];
+            tensor<fp16, [1, 1500, 512]> v_5_cast = linear(bias = var_214_to_fp16, weight = var_213_to_fp16, x = var_194_cast);
+            tensor<int32, [4]> var_222 = const()[name = tensor<string, []>("op_222"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_223_cast = reshape(shape = var_222, x = q_5_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_44_to_fp16 = const()[name = tensor<string, []>("const_44_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> q_7_cast = mul(x = var_223_cast, y = const_44_to_fp16);
+            tensor<int32, [4]> var_229 = const()[name = tensor<string, []>("op_229"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_230_cast = reshape(shape = var_229, x = k_5_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_45_to_fp16 = const()[name = tensor<string, []>("const_45_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> k_7_cast = mul(x = var_230_cast, y = const_45_to_fp16);
+            tensor<int32, [4]> var_236 = const()[name = tensor<string, []>("op_236"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_237_cast = reshape(shape = var_236, x = v_5_cast);
+            tensor<int32, [4]> var_238 = const()[name = tensor<string, []>("op_238"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<bool, []> qk_3_transpose_x_0 = const()[name = tensor<string, []>("qk_3_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> qk_3_transpose_y_0 = const()[name = tensor<string, []>("qk_3_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<int32, [4]> transpose_14_perm_0 = const()[name = tensor<string, []>("transpose_14_perm_0"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [4]> transpose_15_perm_0 = const()[name = tensor<string, []>("transpose_15_perm_0"), val = tensor<int32, [4]>([0, 2, 3, 1])];
+            tensor<fp16, [1, 8, 64, 1500]> transpose_41 = transpose(perm = transpose_15_perm_0, x = k_7_cast);
+            tensor<fp16, [1, 8, 1500, 64]> transpose_42 = transpose(perm = transpose_14_perm_0, x = q_7_cast);
+            tensor<fp16, [1, 8, 1500, 1500]> qk_3_cast = matmul(transpose_x = qk_3_transpose_x_0, transpose_y = qk_3_transpose_y_0, x = transpose_42, y = transpose_41);
+            tensor<fp16, [1, 8, 1500, 1500]> var_242_cast = softmax(axis = var_177, x = qk_3_cast);
+            tensor<bool, []> var_244_transpose_x_0 = const()[name = tensor<string, []>("op_244_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> var_244_transpose_y_0 = const()[name = tensor<string, []>("op_244_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<fp16, [1, 8, 1500, 64]> transpose_43 = transpose(perm = var_238, x = var_237_cast);
+            tensor<fp16, [1, 8, 1500, 64]> var_244_cast = matmul(transpose_x = var_244_transpose_x_0, transpose_y = var_244_transpose_y_0, x = var_242_cast, y = transpose_43);
+            tensor<int32, [4]> var_245 = const()[name = tensor<string, []>("op_245"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [3]> concat_1 = const()[name = tensor<string, []>("concat_1"), val = tensor<int32, [3]>([1, 1500, 512])];
+            tensor<fp16, [1, 1500, 8, 64]> transpose_40 = transpose(perm = var_245, x = var_244_cast);
+            tensor<fp16, [1, 1500, 512]> x_23_cast = reshape(shape = concat_1, x = transpose_40);
+            tensor<fp16, [512, 512]> var_250_to_fp16 = const()[name = tensor<string, []>("op_250_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(11241344)))];
+            tensor<fp16, [512]> var_251_to_fp16 = const()[name = tensor<string, []>("op_251_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(11765696)))];
+            tensor<fp16, [1, 1500, 512]> var_252_cast = linear(bias = var_251_to_fp16, weight = var_250_to_fp16, x = x_23_cast);
+            tensor<fp16, [1, 1500, 512]> x_25_cast = add(x = x_19_cast, y = var_252_cast);
+            tensor<int32, [1]> var_258_axes_0 = const()[name = tensor<string, []>("op_258_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_1_mlp_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_1_mlp_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(11766784)))];
+            tensor<fp16, [512]> blocks_1_mlp_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_1_mlp_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(11767872)))];
+            tensor<fp16, [1, 1500, 512]> var_258_cast = layer_norm(axes = var_258_axes_0, beta = blocks_1_mlp_ln_bias_to_fp16, epsilon = var_183_to_fp16, gamma = blocks_1_mlp_ln_weight_to_fp16, x = x_25_cast);
+            tensor<fp16, [2048, 512]> var_267_to_fp16 = const()[name = tensor<string, []>("op_267_to_fp16"), val = tensor<fp16, [2048, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(11768960)))];
+            tensor<fp16, [2048]> var_268_to_fp16 = const()[name = tensor<string, []>("op_268_to_fp16"), val = tensor<fp16, [2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(13866176)))];
+            tensor<fp16, [1, 1500, 2048]> input_17_cast = linear(bias = var_268_to_fp16, weight = var_267_to_fp16, x = var_258_cast);
+            tensor<string, []> x_29_mode_0 = const()[name = tensor<string, []>("x_29_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 1500, 2048]> x_29_cast = gelu(mode = x_29_mode_0, x = input_17_cast);
+            tensor<fp16, [512, 2048]> var_273_to_fp16 = const()[name = tensor<string, []>("op_273_to_fp16"), val = tensor<fp16, [512, 2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(13870336)))];
+            tensor<fp16, [512]> var_274_to_fp16 = const()[name = tensor<string, []>("op_274_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(15967552)))];
+            tensor<fp16, [1, 1500, 512]> var_275_cast = linear(bias = var_274_to_fp16, weight = var_273_to_fp16, x = x_29_cast);
+            tensor<fp16, [1, 1500, 512]> x_31_cast = add(x = x_25_cast, y = var_275_cast);
+            tensor<int32, []> var_284 = const()[name = tensor<string, []>("op_284"), val = tensor<int32, []>(-1)];
+            tensor<int32, [1]> var_301_axes_0 = const()[name = tensor<string, []>("op_301_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_2_attn_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_2_attn_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(15968640)))];
+            tensor<fp16, [512]> blocks_2_attn_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_2_attn_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(15969728)))];
+            tensor<fp16, []> var_290_to_fp16 = const()[name = tensor<string, []>("op_290_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_301_cast = layer_norm(axes = var_301_axes_0, beta = blocks_2_attn_ln_bias_to_fp16, epsilon = var_290_to_fp16, gamma = blocks_2_attn_ln_weight_to_fp16, x = x_31_cast);
+            tensor<fp16, [512, 512]> var_312_to_fp16 = const()[name = tensor<string, []>("op_312_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(15970816)))];
+            tensor<fp16, [512]> var_313_to_fp16 = const()[name = tensor<string, []>("op_313_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(16495168)))];
+            tensor<fp16, [1, 1500, 512]> q_9_cast = linear(bias = var_313_to_fp16, weight = var_312_to_fp16, x = var_301_cast);
+            tensor<fp16, [512, 512]> var_316_to_fp16 = const()[name = tensor<string, []>("op_316_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(16496256)))];
+            tensor<fp16, [512]> k_9_bias_0_to_fp16 = const()[name = tensor<string, []>("k_9_bias_0_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(17020608)))];
+            tensor<fp16, [1, 1500, 512]> k_9_cast = linear(bias = k_9_bias_0_to_fp16, weight = var_316_to_fp16, x = var_301_cast);
+            tensor<fp16, [512, 512]> var_320_to_fp16 = const()[name = tensor<string, []>("op_320_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(17021696)))];
+            tensor<fp16, [512]> var_321_to_fp16 = const()[name = tensor<string, []>("op_321_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(17546048)))];
+            tensor<fp16, [1, 1500, 512]> v_9_cast = linear(bias = var_321_to_fp16, weight = var_320_to_fp16, x = var_301_cast);
+            tensor<int32, [4]> var_329 = const()[name = tensor<string, []>("op_329"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_330_cast = reshape(shape = var_329, x = q_9_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_46_to_fp16 = const()[name = tensor<string, []>("const_46_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> q_11_cast = mul(x = var_330_cast, y = const_46_to_fp16);
+            tensor<int32, [4]> var_336 = const()[name = tensor<string, []>("op_336"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_337_cast = reshape(shape = var_336, x = k_9_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_47_to_fp16 = const()[name = tensor<string, []>("const_47_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> k_11_cast = mul(x = var_337_cast, y = const_47_to_fp16);
+            tensor<int32, [4]> var_343 = const()[name = tensor<string, []>("op_343"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_344_cast = reshape(shape = var_343, x = v_9_cast);
+            tensor<int32, [4]> var_345 = const()[name = tensor<string, []>("op_345"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<bool, []> qk_5_transpose_x_0 = const()[name = tensor<string, []>("qk_5_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> qk_5_transpose_y_0 = const()[name = tensor<string, []>("qk_5_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<int32, [4]> transpose_16_perm_0 = const()[name = tensor<string, []>("transpose_16_perm_0"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [4]> transpose_17_perm_0 = const()[name = tensor<string, []>("transpose_17_perm_0"), val = tensor<int32, [4]>([0, 2, 3, 1])];
+            tensor<fp16, [1, 8, 64, 1500]> transpose_37 = transpose(perm = transpose_17_perm_0, x = k_11_cast);
+            tensor<fp16, [1, 8, 1500, 64]> transpose_38 = transpose(perm = transpose_16_perm_0, x = q_11_cast);
+            tensor<fp16, [1, 8, 1500, 1500]> qk_5_cast = matmul(transpose_x = qk_5_transpose_x_0, transpose_y = qk_5_transpose_y_0, x = transpose_38, y = transpose_37);
+            tensor<fp16, [1, 8, 1500, 1500]> var_349_cast = softmax(axis = var_284, x = qk_5_cast);
+            tensor<bool, []> var_351_transpose_x_0 = const()[name = tensor<string, []>("op_351_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> var_351_transpose_y_0 = const()[name = tensor<string, []>("op_351_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<fp16, [1, 8, 1500, 64]> transpose_39 = transpose(perm = var_345, x = var_344_cast);
+            tensor<fp16, [1, 8, 1500, 64]> var_351_cast = matmul(transpose_x = var_351_transpose_x_0, transpose_y = var_351_transpose_y_0, x = var_349_cast, y = transpose_39);
+            tensor<int32, [4]> var_352 = const()[name = tensor<string, []>("op_352"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [3]> concat_2 = const()[name = tensor<string, []>("concat_2"), val = tensor<int32, [3]>([1, 1500, 512])];
+            tensor<fp16, [1, 1500, 8, 64]> transpose_36 = transpose(perm = var_352, x = var_351_cast);
+            tensor<fp16, [1, 1500, 512]> x_35_cast = reshape(shape = concat_2, x = transpose_36);
+            tensor<fp16, [512, 512]> var_357_to_fp16 = const()[name = tensor<string, []>("op_357_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(17547136)))];
+            tensor<fp16, [512]> var_358_to_fp16 = const()[name = tensor<string, []>("op_358_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(18071488)))];
+            tensor<fp16, [1, 1500, 512]> var_359_cast = linear(bias = var_358_to_fp16, weight = var_357_to_fp16, x = x_35_cast);
+            tensor<fp16, [1, 1500, 512]> x_37_cast = add(x = x_31_cast, y = var_359_cast);
+            tensor<int32, [1]> var_365_axes_0 = const()[name = tensor<string, []>("op_365_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_2_mlp_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_2_mlp_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(18072576)))];
+            tensor<fp16, [512]> blocks_2_mlp_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_2_mlp_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(18073664)))];
+            tensor<fp16, [1, 1500, 512]> var_365_cast = layer_norm(axes = var_365_axes_0, beta = blocks_2_mlp_ln_bias_to_fp16, epsilon = var_290_to_fp16, gamma = blocks_2_mlp_ln_weight_to_fp16, x = x_37_cast);
+            tensor<fp16, [2048, 512]> var_374_to_fp16 = const()[name = tensor<string, []>("op_374_to_fp16"), val = tensor<fp16, [2048, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(18074752)))];
+            tensor<fp16, [2048]> var_375_to_fp16 = const()[name = tensor<string, []>("op_375_to_fp16"), val = tensor<fp16, [2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(20171968)))];
+            tensor<fp16, [1, 1500, 2048]> input_25_cast = linear(bias = var_375_to_fp16, weight = var_374_to_fp16, x = var_365_cast);
+            tensor<string, []> x_41_mode_0 = const()[name = tensor<string, []>("x_41_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 1500, 2048]> x_41_cast = gelu(mode = x_41_mode_0, x = input_25_cast);
+            tensor<fp16, [512, 2048]> var_380_to_fp16 = const()[name = tensor<string, []>("op_380_to_fp16"), val = tensor<fp16, [512, 2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(20176128)))];
+            tensor<fp16, [512]> var_381_to_fp16 = const()[name = tensor<string, []>("op_381_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(22273344)))];
+            tensor<fp16, [1, 1500, 512]> var_382_cast = linear(bias = var_381_to_fp16, weight = var_380_to_fp16, x = x_41_cast);
+            tensor<fp16, [1, 1500, 512]> x_43_cast = add(x = x_37_cast, y = var_382_cast);
+            tensor<int32, []> var_391 = const()[name = tensor<string, []>("op_391"), val = tensor<int32, []>(-1)];
+            tensor<int32, [1]> var_408_axes_0 = const()[name = tensor<string, []>("op_408_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_3_attn_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_3_attn_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(22274432)))];
+            tensor<fp16, [512]> blocks_3_attn_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_3_attn_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(22275520)))];
+            tensor<fp16, []> var_397_to_fp16 = const()[name = tensor<string, []>("op_397_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_408_cast = layer_norm(axes = var_408_axes_0, beta = blocks_3_attn_ln_bias_to_fp16, epsilon = var_397_to_fp16, gamma = blocks_3_attn_ln_weight_to_fp16, x = x_43_cast);
+            tensor<fp16, [512, 512]> var_419_to_fp16 = const()[name = tensor<string, []>("op_419_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(22276608)))];
+            tensor<fp16, [512]> var_420_to_fp16 = const()[name = tensor<string, []>("op_420_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(22800960)))];
+            tensor<fp16, [1, 1500, 512]> q_13_cast = linear(bias = var_420_to_fp16, weight = var_419_to_fp16, x = var_408_cast);
+            tensor<fp16, [512, 512]> var_423_to_fp16 = const()[name = tensor<string, []>("op_423_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(22802048)))];
+            tensor<fp16, [512]> k_13_bias_0_to_fp16 = const()[name = tensor<string, []>("k_13_bias_0_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(23326400)))];
+            tensor<fp16, [1, 1500, 512]> k_13_cast = linear(bias = k_13_bias_0_to_fp16, weight = var_423_to_fp16, x = var_408_cast);
+            tensor<fp16, [512, 512]> var_427_to_fp16 = const()[name = tensor<string, []>("op_427_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(23327488)))];
+            tensor<fp16, [512]> var_428_to_fp16 = const()[name = tensor<string, []>("op_428_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(23851840)))];
+            tensor<fp16, [1, 1500, 512]> v_13_cast = linear(bias = var_428_to_fp16, weight = var_427_to_fp16, x = var_408_cast);
+            tensor<int32, [4]> var_436 = const()[name = tensor<string, []>("op_436"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_437_cast = reshape(shape = var_436, x = q_13_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_48_to_fp16 = const()[name = tensor<string, []>("const_48_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> q_15_cast = mul(x = var_437_cast, y = const_48_to_fp16);
+            tensor<int32, [4]> var_443 = const()[name = tensor<string, []>("op_443"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_444_cast = reshape(shape = var_443, x = k_13_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_49_to_fp16 = const()[name = tensor<string, []>("const_49_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> k_15_cast = mul(x = var_444_cast, y = const_49_to_fp16);
+            tensor<int32, [4]> var_450 = const()[name = tensor<string, []>("op_450"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_451_cast = reshape(shape = var_450, x = v_13_cast);
+            tensor<int32, [4]> var_452 = const()[name = tensor<string, []>("op_452"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<bool, []> qk_7_transpose_x_0 = const()[name = tensor<string, []>("qk_7_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> qk_7_transpose_y_0 = const()[name = tensor<string, []>("qk_7_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<int32, [4]> transpose_18_perm_0 = const()[name = tensor<string, []>("transpose_18_perm_0"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [4]> transpose_19_perm_0 = const()[name = tensor<string, []>("transpose_19_perm_0"), val = tensor<int32, [4]>([0, 2, 3, 1])];
+            tensor<fp16, [1, 8, 64, 1500]> transpose_33 = transpose(perm = transpose_19_perm_0, x = k_15_cast);
+            tensor<fp16, [1, 8, 1500, 64]> transpose_34 = transpose(perm = transpose_18_perm_0, x = q_15_cast);
+            tensor<fp16, [1, 8, 1500, 1500]> qk_7_cast = matmul(transpose_x = qk_7_transpose_x_0, transpose_y = qk_7_transpose_y_0, x = transpose_34, y = transpose_33);
+            tensor<fp16, [1, 8, 1500, 1500]> var_456_cast = softmax(axis = var_391, x = qk_7_cast);
+            tensor<bool, []> var_458_transpose_x_0 = const()[name = tensor<string, []>("op_458_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> var_458_transpose_y_0 = const()[name = tensor<string, []>("op_458_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<fp16, [1, 8, 1500, 64]> transpose_35 = transpose(perm = var_452, x = var_451_cast);
+            tensor<fp16, [1, 8, 1500, 64]> var_458_cast = matmul(transpose_x = var_458_transpose_x_0, transpose_y = var_458_transpose_y_0, x = var_456_cast, y = transpose_35);
+            tensor<int32, [4]> var_459 = const()[name = tensor<string, []>("op_459"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [3]> concat_3 = const()[name = tensor<string, []>("concat_3"), val = tensor<int32, [3]>([1, 1500, 512])];
+            tensor<fp16, [1, 1500, 8, 64]> transpose_32 = transpose(perm = var_459, x = var_458_cast);
+            tensor<fp16, [1, 1500, 512]> x_47_cast = reshape(shape = concat_3, x = transpose_32);
+            tensor<fp16, [512, 512]> var_464_to_fp16 = const()[name = tensor<string, []>("op_464_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(23852928)))];
+            tensor<fp16, [512]> var_465_to_fp16 = const()[name = tensor<string, []>("op_465_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(24377280)))];
+            tensor<fp16, [1, 1500, 512]> var_466_cast = linear(bias = var_465_to_fp16, weight = var_464_to_fp16, x = x_47_cast);
+            tensor<fp16, [1, 1500, 512]> x_49_cast = add(x = x_43_cast, y = var_466_cast);
+            tensor<int32, [1]> var_472_axes_0 = const()[name = tensor<string, []>("op_472_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_3_mlp_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_3_mlp_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(24378368)))];
+            tensor<fp16, [512]> blocks_3_mlp_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_3_mlp_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(24379456)))];
+            tensor<fp16, [1, 1500, 512]> var_472_cast = layer_norm(axes = var_472_axes_0, beta = blocks_3_mlp_ln_bias_to_fp16, epsilon = var_397_to_fp16, gamma = blocks_3_mlp_ln_weight_to_fp16, x = x_49_cast);
+            tensor<fp16, [2048, 512]> var_481_to_fp16 = const()[name = tensor<string, []>("op_481_to_fp16"), val = tensor<fp16, [2048, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(24380544)))];
+            tensor<fp16, [2048]> var_482_to_fp16 = const()[name = tensor<string, []>("op_482_to_fp16"), val = tensor<fp16, [2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(26477760)))];
+            tensor<fp16, [1, 1500, 2048]> input_33_cast = linear(bias = var_482_to_fp16, weight = var_481_to_fp16, x = var_472_cast);
+            tensor<string, []> x_53_mode_0 = const()[name = tensor<string, []>("x_53_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 1500, 2048]> x_53_cast = gelu(mode = x_53_mode_0, x = input_33_cast);
+            tensor<fp16, [512, 2048]> var_487_to_fp16 = const()[name = tensor<string, []>("op_487_to_fp16"), val = tensor<fp16, [512, 2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(26481920)))];
+            tensor<fp16, [512]> var_488_to_fp16 = const()[name = tensor<string, []>("op_488_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(28579136)))];
+            tensor<fp16, [1, 1500, 512]> var_489_cast = linear(bias = var_488_to_fp16, weight = var_487_to_fp16, x = x_53_cast);
+            tensor<fp16, [1, 1500, 512]> x_55_cast = add(x = x_49_cast, y = var_489_cast);
+            tensor<int32, []> var_498 = const()[name = tensor<string, []>("op_498"), val = tensor<int32, []>(-1)];
+            tensor<int32, [1]> var_515_axes_0 = const()[name = tensor<string, []>("op_515_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_4_attn_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_4_attn_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(28580224)))];
+            tensor<fp16, [512]> blocks_4_attn_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_4_attn_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(28581312)))];
+            tensor<fp16, []> var_504_to_fp16 = const()[name = tensor<string, []>("op_504_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_515_cast = layer_norm(axes = var_515_axes_0, beta = blocks_4_attn_ln_bias_to_fp16, epsilon = var_504_to_fp16, gamma = blocks_4_attn_ln_weight_to_fp16, x = x_55_cast);
+            tensor<fp16, [512, 512]> var_526_to_fp16 = const()[name = tensor<string, []>("op_526_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(28582400)))];
+            tensor<fp16, [512]> var_527_to_fp16 = const()[name = tensor<string, []>("op_527_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(29106752)))];
+            tensor<fp16, [1, 1500, 512]> q_17_cast = linear(bias = var_527_to_fp16, weight = var_526_to_fp16, x = var_515_cast);
+            tensor<fp16, [512, 512]> var_530_to_fp16 = const()[name = tensor<string, []>("op_530_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(29107840)))];
+            tensor<fp16, [512]> k_17_bias_0_to_fp16 = const()[name = tensor<string, []>("k_17_bias_0_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(29632192)))];
+            tensor<fp16, [1, 1500, 512]> k_17_cast = linear(bias = k_17_bias_0_to_fp16, weight = var_530_to_fp16, x = var_515_cast);
+            tensor<fp16, [512, 512]> var_534_to_fp16 = const()[name = tensor<string, []>("op_534_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(29633280)))];
+            tensor<fp16, [512]> var_535_to_fp16 = const()[name = tensor<string, []>("op_535_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(30157632)))];
+            tensor<fp16, [1, 1500, 512]> v_17_cast = linear(bias = var_535_to_fp16, weight = var_534_to_fp16, x = var_515_cast);
+            tensor<int32, [4]> var_543 = const()[name = tensor<string, []>("op_543"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_544_cast = reshape(shape = var_543, x = q_17_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_50_to_fp16 = const()[name = tensor<string, []>("const_50_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> q_19_cast = mul(x = var_544_cast, y = const_50_to_fp16);
+            tensor<int32, [4]> var_550 = const()[name = tensor<string, []>("op_550"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_551_cast = reshape(shape = var_550, x = k_17_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_51_to_fp16 = const()[name = tensor<string, []>("const_51_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> k_19_cast = mul(x = var_551_cast, y = const_51_to_fp16);
+            tensor<int32, [4]> var_557 = const()[name = tensor<string, []>("op_557"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_558_cast = reshape(shape = var_557, x = v_17_cast);
+            tensor<int32, [4]> var_559 = const()[name = tensor<string, []>("op_559"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<bool, []> qk_9_transpose_x_0 = const()[name = tensor<string, []>("qk_9_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> qk_9_transpose_y_0 = const()[name = tensor<string, []>("qk_9_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<int32, [4]> transpose_20_perm_0 = const()[name = tensor<string, []>("transpose_20_perm_0"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [4]> transpose_21_perm_0 = const()[name = tensor<string, []>("transpose_21_perm_0"), val = tensor<int32, [4]>([0, 2, 3, 1])];
+            tensor<fp16, [1, 8, 64, 1500]> transpose_29 = transpose(perm = transpose_21_perm_0, x = k_19_cast);
+            tensor<fp16, [1, 8, 1500, 64]> transpose_30 = transpose(perm = transpose_20_perm_0, x = q_19_cast);
+            tensor<fp16, [1, 8, 1500, 1500]> qk_9_cast = matmul(transpose_x = qk_9_transpose_x_0, transpose_y = qk_9_transpose_y_0, x = transpose_30, y = transpose_29);
+            tensor<fp16, [1, 8, 1500, 1500]> var_563_cast = softmax(axis = var_498, x = qk_9_cast);
+            tensor<bool, []> var_565_transpose_x_0 = const()[name = tensor<string, []>("op_565_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> var_565_transpose_y_0 = const()[name = tensor<string, []>("op_565_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<fp16, [1, 8, 1500, 64]> transpose_31 = transpose(perm = var_559, x = var_558_cast);
+            tensor<fp16, [1, 8, 1500, 64]> var_565_cast = matmul(transpose_x = var_565_transpose_x_0, transpose_y = var_565_transpose_y_0, x = var_563_cast, y = transpose_31);
+            tensor<int32, [4]> var_566 = const()[name = tensor<string, []>("op_566"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [3]> concat_4 = const()[name = tensor<string, []>("concat_4"), val = tensor<int32, [3]>([1, 1500, 512])];
+            tensor<fp16, [1, 1500, 8, 64]> transpose_28 = transpose(perm = var_566, x = var_565_cast);
+            tensor<fp16, [1, 1500, 512]> x_59_cast = reshape(shape = concat_4, x = transpose_28);
+            tensor<fp16, [512, 512]> var_571_to_fp16 = const()[name = tensor<string, []>("op_571_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(30158720)))];
+            tensor<fp16, [512]> var_572_to_fp16 = const()[name = tensor<string, []>("op_572_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(30683072)))];
+            tensor<fp16, [1, 1500, 512]> var_573_cast = linear(bias = var_572_to_fp16, weight = var_571_to_fp16, x = x_59_cast);
+            tensor<fp16, [1, 1500, 512]> x_61_cast = add(x = x_55_cast, y = var_573_cast);
+            tensor<int32, [1]> var_579_axes_0 = const()[name = tensor<string, []>("op_579_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_4_mlp_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_4_mlp_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(30684160)))];
+            tensor<fp16, [512]> blocks_4_mlp_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_4_mlp_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(30685248)))];
+            tensor<fp16, [1, 1500, 512]> var_579_cast = layer_norm(axes = var_579_axes_0, beta = blocks_4_mlp_ln_bias_to_fp16, epsilon = var_504_to_fp16, gamma = blocks_4_mlp_ln_weight_to_fp16, x = x_61_cast);
+            tensor<fp16, [2048, 512]> var_588_to_fp16 = const()[name = tensor<string, []>("op_588_to_fp16"), val = tensor<fp16, [2048, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(30686336)))];
+            tensor<fp16, [2048]> var_589_to_fp16 = const()[name = tensor<string, []>("op_589_to_fp16"), val = tensor<fp16, [2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(32783552)))];
+            tensor<fp16, [1, 1500, 2048]> input_41_cast = linear(bias = var_589_to_fp16, weight = var_588_to_fp16, x = var_579_cast);
+            tensor<string, []> x_65_mode_0 = const()[name = tensor<string, []>("x_65_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 1500, 2048]> x_65_cast = gelu(mode = x_65_mode_0, x = input_41_cast);
+            tensor<fp16, [512, 2048]> var_594_to_fp16 = const()[name = tensor<string, []>("op_594_to_fp16"), val = tensor<fp16, [512, 2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(32787712)))];
+            tensor<fp16, [512]> var_595_to_fp16 = const()[name = tensor<string, []>("op_595_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(34884928)))];
+            tensor<fp16, [1, 1500, 512]> var_596_cast = linear(bias = var_595_to_fp16, weight = var_594_to_fp16, x = x_65_cast);
+            tensor<fp16, [1, 1500, 512]> x_67_cast = add(x = x_61_cast, y = var_596_cast);
+            tensor<int32, []> var_605 = const()[name = tensor<string, []>("op_605"), val = tensor<int32, []>(-1)];
+            tensor<int32, [1]> var_622_axes_0 = const()[name = tensor<string, []>("op_622_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_5_attn_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_5_attn_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(34886016)))];
+            tensor<fp16, [512]> blocks_5_attn_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_5_attn_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(34887104)))];
+            tensor<fp16, []> var_611_to_fp16 = const()[name = tensor<string, []>("op_611_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_622_cast = layer_norm(axes = var_622_axes_0, beta = blocks_5_attn_ln_bias_to_fp16, epsilon = var_611_to_fp16, gamma = blocks_5_attn_ln_weight_to_fp16, x = x_67_cast);
+            tensor<fp16, [512, 512]> var_633_to_fp16 = const()[name = tensor<string, []>("op_633_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(34888192)))];
+            tensor<fp16, [512]> var_634_to_fp16 = const()[name = tensor<string, []>("op_634_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(35412544)))];
+            tensor<fp16, [1, 1500, 512]> q_21_cast = linear(bias = var_634_to_fp16, weight = var_633_to_fp16, x = var_622_cast);
+            tensor<fp16, [512, 512]> var_637_to_fp16 = const()[name = tensor<string, []>("op_637_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(35413632)))];
+            tensor<fp16, [512]> k_21_bias_0_to_fp16 = const()[name = tensor<string, []>("k_21_bias_0_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(35937984)))];
+            tensor<fp16, [1, 1500, 512]> k_21_cast = linear(bias = k_21_bias_0_to_fp16, weight = var_637_to_fp16, x = var_622_cast);
+            tensor<fp16, [512, 512]> var_641_to_fp16 = const()[name = tensor<string, []>("op_641_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(35939072)))];
+            tensor<fp16, [512]> var_642_to_fp16 = const()[name = tensor<string, []>("op_642_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(36463424)))];
+            tensor<fp16, [1, 1500, 512]> v_21_cast = linear(bias = var_642_to_fp16, weight = var_641_to_fp16, x = var_622_cast);
+            tensor<int32, [4]> var_650 = const()[name = tensor<string, []>("op_650"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_651_cast = reshape(shape = var_650, x = q_21_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_52_to_fp16 = const()[name = tensor<string, []>("const_52_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> q_cast = mul(x = var_651_cast, y = const_52_to_fp16);
+            tensor<int32, [4]> var_657 = const()[name = tensor<string, []>("op_657"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_658_cast = reshape(shape = var_657, x = k_21_cast);
+            tensor<fp16, [1, 1, 1, 1]> const_53_to_fp16 = const()[name = tensor<string, []>("const_53_to_fp16"), val = tensor<fp16, [1, 1, 1, 1]>([[[[0x1.6ap-2]]]])];
+            tensor<fp16, [1, 1500, 8, 64]> k_cast = mul(x = var_658_cast, y = const_53_to_fp16);
+            tensor<int32, [4]> var_664 = const()[name = tensor<string, []>("op_664"), val = tensor<int32, [4]>([1, 1500, 8, -1])];
+            tensor<fp16, [1, 1500, 8, 64]> var_665_cast = reshape(shape = var_664, x = v_21_cast);
+            tensor<int32, [4]> var_666 = const()[name = tensor<string, []>("op_666"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<bool, []> qk_transpose_x_0 = const()[name = tensor<string, []>("qk_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> qk_transpose_y_0 = const()[name = tensor<string, []>("qk_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<int32, [4]> transpose_22_perm_0 = const()[name = tensor<string, []>("transpose_22_perm_0"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [4]> transpose_23_perm_0 = const()[name = tensor<string, []>("transpose_23_perm_0"), val = tensor<int32, [4]>([0, 2, 3, 1])];
+            tensor<fp16, [1, 8, 64, 1500]> transpose_25 = transpose(perm = transpose_23_perm_0, x = k_cast);
+            tensor<fp16, [1, 8, 1500, 64]> transpose_26 = transpose(perm = transpose_22_perm_0, x = q_cast);
+            tensor<fp16, [1, 8, 1500, 1500]> qk_cast = matmul(transpose_x = qk_transpose_x_0, transpose_y = qk_transpose_y_0, x = transpose_26, y = transpose_25);
+            tensor<fp16, [1, 8, 1500, 1500]> var_670_cast = softmax(axis = var_605, x = qk_cast);
+            tensor<bool, []> var_672_transpose_x_0 = const()[name = tensor<string, []>("op_672_transpose_x_0"), val = tensor<bool, []>(false)];
+            tensor<bool, []> var_672_transpose_y_0 = const()[name = tensor<string, []>("op_672_transpose_y_0"), val = tensor<bool, []>(false)];
+            tensor<fp16, [1, 8, 1500, 64]> transpose_27 = transpose(perm = var_666, x = var_665_cast);
+            tensor<fp16, [1, 8, 1500, 64]> var_672_cast = matmul(transpose_x = var_672_transpose_x_0, transpose_y = var_672_transpose_y_0, x = var_670_cast, y = transpose_27);
+            tensor<int32, [4]> var_673 = const()[name = tensor<string, []>("op_673"), val = tensor<int32, [4]>([0, 2, 1, 3])];
+            tensor<int32, [3]> concat_5 = const()[name = tensor<string, []>("concat_5"), val = tensor<int32, [3]>([1, 1500, 512])];
+            tensor<fp16, [1, 1500, 8, 64]> transpose_24 = transpose(perm = var_673, x = var_672_cast);
+            tensor<fp16, [1, 1500, 512]> x_71_cast = reshape(shape = concat_5, x = transpose_24);
+            tensor<fp16, [512, 512]> var_678_to_fp16 = const()[name = tensor<string, []>("op_678_to_fp16"), val = tensor<fp16, [512, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(36464512)))];
+            tensor<fp16, [512]> var_679_to_fp16 = const()[name = tensor<string, []>("op_679_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(36988864)))];
+            tensor<fp16, [1, 1500, 512]> var_680_cast = linear(bias = var_679_to_fp16, weight = var_678_to_fp16, x = x_71_cast);
+            tensor<fp16, [1, 1500, 512]> x_73_cast = add(x = x_67_cast, y = var_680_cast);
+            tensor<int32, [1]> var_686_axes_0 = const()[name = tensor<string, []>("op_686_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> blocks_5_mlp_ln_weight_to_fp16 = const()[name = tensor<string, []>("blocks_5_mlp_ln_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(36989952)))];
+            tensor<fp16, [512]> blocks_5_mlp_ln_bias_to_fp16 = const()[name = tensor<string, []>("blocks_5_mlp_ln_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(36991040)))];
+            tensor<fp16, [1, 1500, 512]> var_686_cast = layer_norm(axes = var_686_axes_0, beta = blocks_5_mlp_ln_bias_to_fp16, epsilon = var_611_to_fp16, gamma = blocks_5_mlp_ln_weight_to_fp16, x = x_73_cast);
+            tensor<fp16, [2048, 512]> var_695_to_fp16 = const()[name = tensor<string, []>("op_695_to_fp16"), val = tensor<fp16, [2048, 512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(36992128)))];
+            tensor<fp16, [2048]> var_696_to_fp16 = const()[name = tensor<string, []>("op_696_to_fp16"), val = tensor<fp16, [2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(39089344)))];
+            tensor<fp16, [1, 1500, 2048]> input_49_cast = linear(bias = var_696_to_fp16, weight = var_695_to_fp16, x = var_686_cast);
+            tensor<string, []> x_77_mode_0 = const()[name = tensor<string, []>("x_77_mode_0"), val = tensor<string, []>("EXACT")];
+            tensor<fp16, [1, 1500, 2048]> x_77_cast = gelu(mode = x_77_mode_0, x = input_49_cast);
+            tensor<fp16, [512, 2048]> var_701_to_fp16 = const()[name = tensor<string, []>("op_701_to_fp16"), val = tensor<fp16, [512, 2048]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(39093504)))];
+            tensor<fp16, [512]> var_702_to_fp16 = const()[name = tensor<string, []>("op_702_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(41190720)))];
+            tensor<fp16, [1, 1500, 512]> var_703_cast = linear(bias = var_702_to_fp16, weight = var_701_to_fp16, x = x_77_cast);
+            tensor<fp16, [1, 1500, 512]> x_cast = add(x = x_73_cast, y = var_703_cast);
+            tensor<int32, [1]> var_716_axes_0 = const()[name = tensor<string, []>("op_716_axes_0"), val = tensor<int32, [1]>([-1])];
+            tensor<fp16, [512]> ln_post_weight_to_fp16 = const()[name = tensor<string, []>("ln_post_weight_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(41191808)))];
+            tensor<fp16, [512]> ln_post_bias_to_fp16 = const()[name = tensor<string, []>("ln_post_bias_to_fp16"), val = tensor<fp16, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(41192896)))];
+            tensor<fp16, []> var_707_to_fp16 = const()[name = tensor<string, []>("op_707_to_fp16"), val = tensor<fp16, []>(0x1.5p-17)];
+            tensor<fp16, [1, 1500, 512]> var_716_cast = layer_norm(axes = var_716_axes_0, beta = ln_post_bias_to_fp16, epsilon = var_707_to_fp16, gamma = ln_post_weight_to_fp16, x = x_cast);
+            tensor<string, []> var_716_cast_to_fp32_dtype_0 = const()[name = tensor<string, []>("op_716_cast_to_fp32_dtype_0"), val = tensor<string, []>("fp32")];
+            tensor<fp32, [1, 1500, 512]> output = cast(dtype = var_716_cast_to_fp32_dtype_0, x = var_716_cast);
+        } -> (output);
+}

ggml-base.en-encoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1578c2b8925c0ff867a8bb96dd56c0eb75df3b909ca9507dc2aa1e437c296dc9
+size 41193984

ggml-large-v1-encoder.mlmodelc.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d10d24a4272572fee07843ffc2e4e4ceebd1d0b0831a142dd2fdb3d56d5bf56e
-size 1177529527
+oid sha256:205f888f1f31bd06ff6650bec836e332e6f84f28dfbc2c00378559162b3584f4
+size 1177530049

ggml-large-v1-encoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9b73481ce4a6af49154956c7f9c72dc02d96448fa69651f5c5a2124edfc8e31
+size 243

ggml-large-v1-encoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05fe28591b40616fa0c34ad7b853133623f5300923ec812acb11459c411acf3b
+size 149

ggml-large-v1-encoder.mlmodelc/metadata.json

@@ -0,0 +1,65 @@
+[
+  {
+    "metadataOutputVersion" : "3.0",
+    "storagePrecision" : "Float16",
+    "outputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32)",
+        "shortDescription" : "",
+        "shape" : "[]",
+        "name" : "output",
+        "type" : "MultiArray"
+      }
+    ],
+    "modelParameters" : [
+
+    ],
+    "specificationVersion" : 6,
+    "mlProgramOperationTypeHistogram" : {
+      "Linear" : 192,
+      "Matmul" : 64,
+      "Cast" : 2,
+      "Conv" : 2,
+      "Softmax" : 32,
+      "Add" : 65,
+      "LayerNorm" : 65,
+      "Mul" : 64,
+      "Transpose" : 129,
+      "Gelu" : 34,
+      "Reshape" : 128
+    },
+    "computePrecision" : "Mixed (Float16, Float32, Int32)",
+    "isUpdatable" : "0",
+    "availability" : {
+      "macOS" : "12.0",
+      "tvOS" : "15.0",
+      "visionOS" : "1.0",
+      "watchOS" : "8.0",
+      "iOS" : "15.0",
+      "macCatalyst" : "15.0"
+    },
+    "modelType" : {
+      "name" : "MLModelType_mlProgram"
+    },
+    "userDefinedMetadata" : {
+
+    },
+    "inputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32 1 × 80 × 3000)",
+        "shortDescription" : "",
+        "shape" : "[1, 80, 3000]",
+        "name" : "logmel_data",
+        "type" : "MultiArray"
+      }
+    ],
+    "generatedClassName" : "coreml_encoder_large_v1",
+    "method" : "predict"
+  }
+]

ggml-large-v1-encoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe6163af221e89e96ebdd4163bddbf79670be7ee6aac34f1d662ce42f6fb014d
+size 1273684480

ggml-large-v2-encoder.mlmodelc.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c65d28737f51d09dff3b9da9d26c2ab6f5d82857c9b049ce383d64c2502a5541
-size 1174643458
+oid sha256:08d4306667aa7378a8ed9ecbdc94337fe01a2d71c9afe40c4e8dfb967f7bfd41
+size 1174643634

ggml-large-v2-encoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9b73481ce4a6af49154956c7f9c72dc02d96448fa69651f5c5a2124edfc8e31
+size 243

ggml-large-v2-encoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05fe28591b40616fa0c34ad7b853133623f5300923ec812acb11459c411acf3b
+size 149

ggml-large-v2-encoder.mlmodelc/metadata.json

@@ -0,0 +1,65 @@
+[
+  {
+    "metadataOutputVersion" : "3.0",
+    "storagePrecision" : "Float16",
+    "outputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32)",
+        "shortDescription" : "",
+        "shape" : "[]",
+        "name" : "output",
+        "type" : "MultiArray"
+      }
+    ],
+    "modelParameters" : [
+
+    ],
+    "specificationVersion" : 6,
+    "mlProgramOperationTypeHistogram" : {
+      "Linear" : 192,
+      "Matmul" : 64,
+      "Cast" : 2,
+      "Conv" : 2,
+      "Softmax" : 32,
+      "Add" : 65,
+      "LayerNorm" : 65,
+      "Mul" : 64,
+      "Transpose" : 129,
+      "Gelu" : 34,
+      "Reshape" : 128
+    },
+    "computePrecision" : "Mixed (Float16, Float32, Int32)",
+    "isUpdatable" : "0",
+    "availability" : {
+      "macOS" : "12.0",
+      "tvOS" : "15.0",
+      "visionOS" : "1.0",
+      "watchOS" : "8.0",
+      "iOS" : "15.0",
+      "macCatalyst" : "15.0"
+    },
+    "modelType" : {
+      "name" : "MLModelType_mlProgram"
+    },
+    "userDefinedMetadata" : {
+
+    },
+    "inputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32 1 × 80 × 3000)",
+        "shortDescription" : "",
+        "shape" : "[1, 80, 3000]",
+        "name" : "logmel_data",
+        "type" : "MultiArray"
+      }
+    ],
+    "generatedClassName" : "coreml_encoder_large_v2",
+    "method" : "predict"
+  }
+]

ggml-large-v2-encoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42f92759b6f55ec227a3cb7e042c0424219cc781818ed3b8b0e72743714ed5d7
+size 1273684480

ggml-large-v3-encoder.mlmodelc.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9be410daad332c1756c34aa119479ef0b0de80c770b094da4ab5cc29ffa417ee
+size 1174643677

ggml-large-v3-encoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9b73481ce4a6af49154956c7f9c72dc02d96448fa69651f5c5a2124edfc8e31
+size 243

ggml-large-v3-encoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05fe28591b40616fa0c34ad7b853133623f5300923ec812acb11459c411acf3b
+size 149

ggml-large-v3-encoder.mlmodelc/metadata.json

@@ -0,0 +1,65 @@
+[
+  {
+    "metadataOutputVersion" : "3.0",
+    "storagePrecision" : "Float16",
+    "outputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32)",
+        "shortDescription" : "",
+        "shape" : "[]",
+        "name" : "output",
+        "type" : "MultiArray"
+      }
+    ],
+    "modelParameters" : [
+
+    ],
+    "specificationVersion" : 6,
+    "mlProgramOperationTypeHistogram" : {
+      "Linear" : 192,
+      "Matmul" : 64,
+      "Cast" : 2,
+      "Conv" : 2,
+      "Softmax" : 32,
+      "Add" : 65,
+      "LayerNorm" : 65,
+      "Mul" : 64,
+      "Transpose" : 129,
+      "Gelu" : 34,
+      "Reshape" : 128
+    },
+    "computePrecision" : "Mixed (Float16, Float32, Int32)",
+    "isUpdatable" : "0",
+    "availability" : {
+      "macOS" : "12.0",
+      "tvOS" : "15.0",
+      "visionOS" : "1.0",
+      "watchOS" : "8.0",
+      "iOS" : "15.0",
+      "macCatalyst" : "15.0"
+    },
+    "modelType" : {
+      "name" : "MLModelType_mlProgram"
+    },
+    "userDefinedMetadata" : {
+
+    },
+    "inputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32 1 × 80 × 3000)",
+        "shortDescription" : "",
+        "shape" : "[1, 80, 3000]",
+        "name" : "logmel_data",
+        "type" : "MultiArray"
+      }
+    ],
+    "generatedClassName" : "coreml_encoder_large",
+    "method" : "predict"
+  }
+]

ggml-large-v3-encoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3c79ca8b3de455956bb7e2af31f525413431950048a54b2e4c72c39d8c5acc7
+size 1273684480

ggml-medium-encoder.mlmodelc.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:79b0b8d436d47d3f24dd3afc91f19447dd686a4f37521b2f6d9c30a642133fbd
+oid sha256:991295c77a7f73bd04b9310e8ca2aa279826acb7c643f7973cd27fa892487377
 size 567829413

ggml-medium-encoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:adbe456375e7eb3407732a426ecb65bbda86860e4aa801f3a696b70b8a533cdd
+size 207

ggml-medium-encoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05fe28591b40616fa0c34ad7b853133623f5300923ec812acb11459c411acf3b
+size 149

ggml-medium-encoder.mlmodelc/metadata.json

@@ -0,0 +1,64 @@
+[
+  {
+    "metadataOutputVersion" : "3.0",
+    "storagePrecision" : "Float16",
+    "outputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32)",
+        "shortDescription" : "",
+        "shape" : "[]",
+        "name" : "output",
+        "type" : "MultiArray"
+      }
+    ],
+    "modelParameters" : [
+
+    ],
+    "specificationVersion" : 6,
+    "mlProgramOperationTypeHistogram" : {
+      "Linear" : 144,
+      "Matmul" : 48,
+      "Cast" : 2,
+      "Conv" : 2,
+      "Softmax" : 24,
+      "Add" : 49,
+      "LayerNorm" : 49,
+      "Mul" : 48,
+      "Transpose" : 97,
+      "Gelu" : 26,
+      "Reshape" : 96
+    },
+    "computePrecision" : "Mixed (Float16, Float32, Int32)",
+    "isUpdatable" : "0",
+    "availability" : {
+      "macOS" : "12.0",
+      "tvOS" : "15.0",
+      "watchOS" : "8.0",
+      "iOS" : "15.0",
+      "macCatalyst" : "15.0"
+    },
+    "modelType" : {
+      "name" : "MLModelType_mlProgram"
+    },
+    "userDefinedMetadata" : {
+
+    },
+    "inputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32 1 × 80 × 3000)",
+        "shortDescription" : "",
+        "shape" : "[1, 80, 3000]",
+        "name" : "logmel_data",
+        "type" : "MultiArray"
+      }
+    ],
+    "generatedClassName" : "coreml_encoder_medium",
+    "method" : "predict"
+  }
+]

ggml-medium-encoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a188b0e4e3109f28f38f1f47ea2497ffe623923419df8e1ae12cb5f809a1815
+size 614507008

ggml-medium.en-encoder.mlmodelc.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cdc44fee3c62b5743913e3147ed75f4e8ecfb52dd7a0f0f7387094b406ff0ee6
+oid sha256:57cf30e4542de7ce7b798573d99e3198ebc435c2fd071672f9fd5c71d2984b99
 size 566993085

ggml-medium.en-encoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:adbe456375e7eb3407732a426ecb65bbda86860e4aa801f3a696b70b8a533cdd
+size 207

ggml-medium.en-encoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05fe28591b40616fa0c34ad7b853133623f5300923ec812acb11459c411acf3b
+size 149

ggml-medium.en-encoder.mlmodelc/metadata.json

@@ -0,0 +1,64 @@
+[
+  {
+    "metadataOutputVersion" : "3.0",
+    "storagePrecision" : "Float16",
+    "outputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32)",
+        "shortDescription" : "",
+        "shape" : "[]",
+        "name" : "output",
+        "type" : "MultiArray"
+      }
+    ],
+    "modelParameters" : [
+
+    ],
+    "specificationVersion" : 6,
+    "mlProgramOperationTypeHistogram" : {
+      "Linear" : 144,
+      "Matmul" : 48,
+      "Cast" : 2,
+      "Conv" : 2,
+      "Softmax" : 24,
+      "Add" : 49,
+      "LayerNorm" : 49,
+      "Mul" : 48,
+      "Transpose" : 97,
+      "Gelu" : 26,
+      "Reshape" : 96
+    },
+    "computePrecision" : "Mixed (Float16, Float32, Int32)",
+    "isUpdatable" : "0",
+    "availability" : {
+      "macOS" : "12.0",
+      "tvOS" : "15.0",
+      "watchOS" : "8.0",
+      "iOS" : "15.0",
+      "macCatalyst" : "15.0"
+    },
+    "modelType" : {
+      "name" : "MLModelType_mlProgram"
+    },
+    "userDefinedMetadata" : {
+
+    },
+    "inputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32 1 × 80 × 3000)",
+        "shortDescription" : "",
+        "shape" : "[1, 80, 3000]",
+        "name" : "logmel_data",
+        "type" : "MultiArray"
+      }
+    ],
+    "generatedClassName" : "coreml_encoder_medium_en",
+    "method" : "predict"
+  }
+]

ggml-medium.en-encoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63a6c6a2c6cb26aaca930ccc03fe96e2e5820d1ae2802010900eff36b226674d
+size 614507008

ggml-small-encoder.mlmodelc.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de43fb9fed471e95c19e60ae67575c2bf09e8fb607016da171b06ddad313988b
+oid sha256:c8b65b12409d1fd32029d42992903f4e5aef6679da3a4835bcb85738a19e1057
 size 163083239

ggml-small-encoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18ad2072ae82872c2ba8a187071e1e7d6c1105253685e7aa95138adcf07874e0
+size 207

ggml-small-encoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05fe28591b40616fa0c34ad7b853133623f5300923ec812acb11459c411acf3b
+size 149

ggml-small-encoder.mlmodelc/metadata.json

@@ -0,0 +1,64 @@
+[
+  {
+    "metadataOutputVersion" : "3.0",
+    "storagePrecision" : "Float16",
+    "outputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32)",
+        "shortDescription" : "",
+        "shape" : "[]",
+        "name" : "output",
+        "type" : "MultiArray"
+      }
+    ],
+    "modelParameters" : [
+
+    ],
+    "specificationVersion" : 6,
+    "mlProgramOperationTypeHistogram" : {
+      "Linear" : 72,
+      "Matmul" : 24,
+      "Cast" : 2,
+      "Conv" : 2,
+      "Softmax" : 12,
+      "Add" : 25,
+      "LayerNorm" : 25,
+      "Mul" : 24,
+      "Transpose" : 49,
+      "Gelu" : 14,
+      "Reshape" : 48
+    },
+    "computePrecision" : "Mixed (Float16, Float32, Int32)",
+    "isUpdatable" : "0",
+    "availability" : {
+      "macOS" : "12.0",
+      "tvOS" : "15.0",
+      "watchOS" : "8.0",
+      "iOS" : "15.0",
+      "macCatalyst" : "15.0"
+    },
+    "modelType" : {
+      "name" : "MLModelType_mlProgram"
+    },
+    "userDefinedMetadata" : {
+
+    },
+    "inputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32 1 × 80 × 3000)",
+        "shortDescription" : "",
+        "shape" : "[1, 80, 3000]",
+        "name" : "logmel_data",
+        "type" : "MultiArray"
+      }
+    ],
+    "generatedClassName" : "coreml_encoder_small",
+    "method" : "predict"
+  }
+]

ggml-small-encoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87eed4ae76f11a2d4a50786bc7423d4b45c2d0d9ca05577a3bd2557452072eaf
+size 176339456

ggml-small.en-encoder.mlmodelc.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2ef1c506378b825b4b4341979a93e1656b5d6c129f17114cfb8fb78aabc2f89
+oid sha256:677fad005c458eba9a1d69a8e4a0bef1b129b208585ca90a4dd8ba4f8610eaed
 size 162952446