nexaml commited on
Commit
0c837d4
·
verified ·
1 Parent(s): 45564a5

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +1 -0
  2. qwen3-4B-part0-ar1-32.mlmodelc/analytics/coremldata.bin +3 -0
  3. qwen3-4B-part0-ar1-32.mlmodelc/coremldata.bin +3 -0
  4. qwen3-4B-part0-ar1-32.mlmodelc/metadata.json +350 -0
  5. qwen3-4B-part0-ar1-32.mlmodelc/model.mil +0 -0
  6. qwen3-4B-part0-ar1-32.mlmodelc/weights/weight.bin +3 -0
  7. qwen3-4B-part1-ar1-32.mlmodelc/analytics/coremldata.bin +3 -0
  8. qwen3-4B-part1-ar1-32.mlmodelc/coremldata.bin +3 -0
  9. qwen3-4B-part1-ar1-32.mlmodelc/metadata.json +350 -0
  10. qwen3-4B-part1-ar1-32.mlmodelc/model.mil +0 -0
  11. qwen3-4B-part1-ar1-32.mlmodelc/weights/weight.bin +3 -0
  12. qwen3-4B-part2-ar1-32.mlmodelc/analytics/coremldata.bin +3 -0
  13. qwen3-4B-part2-ar1-32.mlmodelc/coremldata.bin +3 -0
  14. qwen3-4B-part2-ar1-32.mlmodelc/metadata.json +350 -0
  15. qwen3-4B-part2-ar1-32.mlmodelc/model.mil +0 -0
  16. qwen3-4B-part2-ar1-32.mlmodelc/weights/weight.bin +3 -0
  17. qwen3-4B-part3-ar1-32.mlmodelc/analytics/coremldata.bin +3 -0
  18. qwen3-4B-part3-ar1-32.mlmodelc/coremldata.bin +3 -0
  19. qwen3-4B-part3-ar1-32.mlmodelc/metadata.json +350 -0
  20. qwen3-4B-part3-ar1-32.mlmodelc/model.mil +0 -0
  21. qwen3-4B-part3-ar1-32.mlmodelc/weights/weight.bin +3 -0
  22. qwen3-4B-part4-ar1-32.mlmodelc/analytics/coremldata.bin +3 -0
  23. qwen3-4B-part4-ar1-32.mlmodelc/coremldata.bin +3 -0
  24. qwen3-4B-part4-ar1-32.mlmodelc/metadata.json +350 -0
  25. qwen3-4B-part4-ar1-32.mlmodelc/model.mil +0 -0
  26. qwen3-4B-part4-ar1-32.mlmodelc/weights/weight.bin +3 -0
  27. qwen3-4B-part5-ar1-32.mlmodelc/analytics/coremldata.bin +3 -0
  28. qwen3-4B-part5-ar1-32.mlmodelc/coremldata.bin +3 -0
  29. qwen3-4B-part5-ar1-32.mlmodelc/metadata.json +350 -0
  30. qwen3-4B-part5-ar1-32.mlmodelc/model.mil +0 -0
  31. qwen3-4B-part5-ar1-32.mlmodelc/weights/weight.bin +3 -0
  32. qwen3-4B-part6-ar1-32.mlmodelc/analytics/coremldata.bin +3 -0
  33. qwen3-4B-part6-ar1-32.mlmodelc/coremldata.bin +3 -0
  34. qwen3-4B-part6-ar1-32.mlmodelc/metadata.json +350 -0
  35. qwen3-4B-part6-ar1-32.mlmodelc/model.mil +0 -0
  36. qwen3-4B-part6-ar1-32.mlmodelc/weights/weight.bin +3 -0
  37. qwen3-4B-part7-ar1-32.mlmodelc/analytics/coremldata.bin +3 -0
  38. qwen3-4B-part7-ar1-32.mlmodelc/coremldata.bin +3 -0
  39. qwen3-4B-part7-ar1-32.mlmodelc/metadata.json +350 -0
  40. qwen3-4B-part7-ar1-32.mlmodelc/model.mil +0 -0
  41. qwen3-4B-part7-ar1-32.mlmodelc/weights/weight.bin +3 -0
  42. qwen3-4B-part8-ar1-32.mlmodelc/analytics/coremldata.bin +3 -0
  43. qwen3-4B-part8-ar1-32.mlmodelc/coremldata.bin +3 -0
  44. qwen3-4B-part8-ar1-32.mlmodelc/metadata.json +350 -0
  45. qwen3-4B-part8-ar1-32.mlmodelc/model.mil +0 -0
  46. qwen3-4B-part8-ar1-32.mlmodelc/weights/weight.bin +3 -0
  47. utils/added_tokens.json +28 -0
  48. utils/chat_template.jinja +89 -0
  49. utils/config.json +68 -0
  50. utils/embed.pt +3 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ utils/tokenizer.json filter=lfs diff=lfs merge=lfs -text
qwen3-4B-part0-ar1-32.mlmodelc/analytics/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:846768a9d33810840860dbf42fd49f635d2f5ad3fdb34610e80e276241f7dd90
3
+ size 243
qwen3-4B-part0-ar1-32.mlmodelc/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6181f4eb9eec7d084dc9b7c68b9747a75634e25e74b39b9c3aa26dc526d2c5c
3
+ size 577
qwen3-4B-part0-ar1-32.mlmodelc/metadata.json ADDED
@@ -0,0 +1,350 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "metadataOutputVersion" : "3.0",
4
+ "storagePrecision" : "Float16",
5
+ "outputSchema" : [
6
+ {
7
+ "hasShapeFlexibility" : "0",
8
+ "isOptional" : "0",
9
+ "dataType" : "Float16",
10
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
11
+ "shortDescription" : "",
12
+ "shape" : "[1, 1, 2560]",
13
+ "name" : "output",
14
+ "type" : "MultiArray"
15
+ }
16
+ ],
17
+ "modelParameters" : [
18
+
19
+ ],
20
+ "specificationVersion" : 9,
21
+ "functions" : [
22
+ {
23
+ "inputSchema" : [
24
+ {
25
+ "hasShapeFlexibility" : "0",
26
+ "isOptional" : "0",
27
+ "dataType" : "Float16",
28
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
29
+ "shortDescription" : "",
30
+ "shape" : "[1, 1, 2560]",
31
+ "name" : "input_embedding",
32
+ "type" : "MultiArray"
33
+ },
34
+ {
35
+ "hasShapeFlexibility" : "0",
36
+ "isOptional" : "0",
37
+ "dataType" : "Float16",
38
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
39
+ "shortDescription" : "",
40
+ "shape" : "[1, 1, 128]",
41
+ "name" : "cos",
42
+ "type" : "MultiArray"
43
+ },
44
+ {
45
+ "hasShapeFlexibility" : "0",
46
+ "isOptional" : "0",
47
+ "dataType" : "Float16",
48
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
49
+ "shortDescription" : "",
50
+ "shape" : "[1, 1, 128]",
51
+ "name" : "sin",
52
+ "type" : "MultiArray"
53
+ },
54
+ {
55
+ "hasShapeFlexibility" : "0",
56
+ "isOptional" : "0",
57
+ "dataType" : "Float16",
58
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
59
+ "shortDescription" : "",
60
+ "shape" : "[1, 1, 1, 1024]",
61
+ "name" : "attention_mask",
62
+ "type" : "MultiArray"
63
+ },
64
+ {
65
+ "hasShapeFlexibility" : "0",
66
+ "isOptional" : "0",
67
+ "dataType" : "Int32",
68
+ "formattedType" : "MultiArray (Int32 1)",
69
+ "shortDescription" : "",
70
+ "shape" : "[1]",
71
+ "name" : "n_past",
72
+ "type" : "MultiArray"
73
+ }
74
+ ],
75
+ "computePrecision" : "Mixed (Float16, Int32)",
76
+ "storagePrecision" : "Float16",
77
+ "stateSchema" : [
78
+ {
79
+ "dataType" : "Float16",
80
+ "isOptional" : "0",
81
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
82
+ "shortDescription" : "",
83
+ "shape" : "[4, 8, 1024, 128]",
84
+ "name" : "key_cache",
85
+ "type" : "State"
86
+ },
87
+ {
88
+ "dataType" : "Float16",
89
+ "isOptional" : "0",
90
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
91
+ "shortDescription" : "",
92
+ "shape" : "[4, 8, 1024, 128]",
93
+ "name" : "value_cache",
94
+ "type" : "State"
95
+ }
96
+ ],
97
+ "outputSchema" : [
98
+ {
99
+ "hasShapeFlexibility" : "0",
100
+ "isOptional" : "0",
101
+ "dataType" : "Float16",
102
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
103
+ "shortDescription" : "",
104
+ "shape" : "[1, 1, 2560]",
105
+ "name" : "output",
106
+ "type" : "MultiArray"
107
+ }
108
+ ],
109
+ "name" : "ar1",
110
+ "mlProgramOperationTypeHistogram" : {
111
+ "Ios18.readState" : 10,
112
+ "Ios18.conv" : 28,
113
+ "Ios18.expandDims" : 26,
114
+ "Ios18.matmul" : 8,
115
+ "Ios18.concat" : 29,
116
+ "Ios18.add" : 21,
117
+ "Ios18.silu" : 4,
118
+ "Ios18.sliceUpdate" : 8,
119
+ "Ios18.sliceByIndex" : 40,
120
+ "Ios18.layerNorm" : 16,
121
+ "Ios18.transpose" : 36,
122
+ "Tile" : 8,
123
+ "Ios18.writeState" : 8,
124
+ "Ios18.reshape" : 32,
125
+ "Ios18.softmax" : 4,
126
+ "Ios18.squeeze" : 4,
127
+ "Ios18.mul" : 64
128
+ }
129
+ },
130
+ {
131
+ "inputSchema" : [
132
+ {
133
+ "hasShapeFlexibility" : "0",
134
+ "isOptional" : "0",
135
+ "dataType" : "Float16",
136
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
137
+ "shortDescription" : "",
138
+ "shape" : "[1, 32, 2560]",
139
+ "name" : "input_embedding",
140
+ "type" : "MultiArray"
141
+ },
142
+ {
143
+ "hasShapeFlexibility" : "0",
144
+ "isOptional" : "0",
145
+ "dataType" : "Float16",
146
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
147
+ "shortDescription" : "",
148
+ "shape" : "[1, 32, 128]",
149
+ "name" : "cos",
150
+ "type" : "MultiArray"
151
+ },
152
+ {
153
+ "hasShapeFlexibility" : "0",
154
+ "isOptional" : "0",
155
+ "dataType" : "Float16",
156
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
157
+ "shortDescription" : "",
158
+ "shape" : "[1, 32, 128]",
159
+ "name" : "sin",
160
+ "type" : "MultiArray"
161
+ },
162
+ {
163
+ "hasShapeFlexibility" : "0",
164
+ "isOptional" : "0",
165
+ "dataType" : "Float16",
166
+ "formattedType" : "MultiArray (Float16 1 × 1 × 32 × 1024)",
167
+ "shortDescription" : "",
168
+ "shape" : "[1, 1, 32, 1024]",
169
+ "name" : "attention_mask",
170
+ "type" : "MultiArray"
171
+ },
172
+ {
173
+ "hasShapeFlexibility" : "0",
174
+ "isOptional" : "0",
175
+ "dataType" : "Int32",
176
+ "formattedType" : "MultiArray (Int32 1)",
177
+ "shortDescription" : "",
178
+ "shape" : "[1]",
179
+ "name" : "n_past",
180
+ "type" : "MultiArray"
181
+ }
182
+ ],
183
+ "computePrecision" : "Mixed (Float16, Int32)",
184
+ "storagePrecision" : "Float16",
185
+ "stateSchema" : [
186
+ {
187
+ "dataType" : "Float16",
188
+ "isOptional" : "0",
189
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
190
+ "shortDescription" : "",
191
+ "shape" : "[4, 8, 1024, 128]",
192
+ "name" : "key_cache",
193
+ "type" : "State"
194
+ },
195
+ {
196
+ "dataType" : "Float16",
197
+ "isOptional" : "0",
198
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
199
+ "shortDescription" : "",
200
+ "shape" : "[4, 8, 1024, 128]",
201
+ "name" : "value_cache",
202
+ "type" : "State"
203
+ }
204
+ ],
205
+ "outputSchema" : [
206
+ {
207
+ "hasShapeFlexibility" : "0",
208
+ "isOptional" : "0",
209
+ "dataType" : "Float16",
210
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
211
+ "shortDescription" : "",
212
+ "shape" : "[1, 32, 2560]",
213
+ "name" : "output",
214
+ "type" : "MultiArray"
215
+ }
216
+ ],
217
+ "name" : "ar32",
218
+ "mlProgramOperationTypeHistogram" : {
219
+ "Ios18.readState" : 10,
220
+ "Ios18.conv" : 28,
221
+ "Ios18.expandDims" : 26,
222
+ "Ios18.matmul" : 8,
223
+ "Ios18.concat" : 29,
224
+ "Ios18.add" : 21,
225
+ "Ios18.silu" : 4,
226
+ "Ios18.sliceUpdate" : 8,
227
+ "Ios18.sliceByIndex" : 40,
228
+ "Ios18.layerNorm" : 16,
229
+ "Ios18.transpose" : 36,
230
+ "Tile" : 8,
231
+ "Ios18.writeState" : 8,
232
+ "Ios18.reshape" : 32,
233
+ "Ios18.softmax" : 4,
234
+ "Ios18.squeeze" : 4,
235
+ "Ios18.mul" : 64
236
+ }
237
+ }
238
+ ],
239
+ "mlProgramOperationTypeHistogram" : {
240
+ "Ios18.readState" : 10,
241
+ "Ios18.conv" : 28,
242
+ "Ios18.expandDims" : 26,
243
+ "Ios18.matmul" : 8,
244
+ "Ios18.concat" : 29,
245
+ "Ios18.add" : 21,
246
+ "Ios18.silu" : 4,
247
+ "Ios18.sliceUpdate" : 8,
248
+ "Ios18.sliceByIndex" : 40,
249
+ "Ios18.layerNorm" : 16,
250
+ "Ios18.transpose" : 36,
251
+ "Tile" : 8,
252
+ "Ios18.writeState" : 8,
253
+ "Ios18.reshape" : 32,
254
+ "Ios18.softmax" : 4,
255
+ "Ios18.squeeze" : 4,
256
+ "Ios18.mul" : 64
257
+ },
258
+ "isUpdatable" : "0",
259
+ "stateSchema" : [
260
+ {
261
+ "dataType" : "Float16",
262
+ "isOptional" : "0",
263
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
264
+ "shortDescription" : "",
265
+ "shape" : "[4, 8, 1024, 128]",
266
+ "name" : "key_cache",
267
+ "type" : "State"
268
+ },
269
+ {
270
+ "dataType" : "Float16",
271
+ "isOptional" : "0",
272
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
273
+ "shortDescription" : "",
274
+ "shape" : "[4, 8, 1024, 128]",
275
+ "name" : "value_cache",
276
+ "type" : "State"
277
+ }
278
+ ],
279
+ "availability" : {
280
+ "macOS" : "15.0",
281
+ "tvOS" : "18.0",
282
+ "visionOS" : "2.0",
283
+ "watchOS" : "11.0",
284
+ "iOS" : "18.0",
285
+ "macCatalyst" : "18.0"
286
+ },
287
+ "computePrecision" : "Mixed (Float16, Int32)",
288
+ "modelType" : {
289
+ "name" : "MLModelType_mlProgram"
290
+ },
291
+ "inputSchema" : [
292
+ {
293
+ "hasShapeFlexibility" : "0",
294
+ "isOptional" : "0",
295
+ "dataType" : "Float16",
296
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
297
+ "shortDescription" : "",
298
+ "shape" : "[1, 1, 2560]",
299
+ "name" : "input_embedding",
300
+ "type" : "MultiArray"
301
+ },
302
+ {
303
+ "hasShapeFlexibility" : "0",
304
+ "isOptional" : "0",
305
+ "dataType" : "Float16",
306
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
307
+ "shortDescription" : "",
308
+ "shape" : "[1, 1, 128]",
309
+ "name" : "cos",
310
+ "type" : "MultiArray"
311
+ },
312
+ {
313
+ "hasShapeFlexibility" : "0",
314
+ "isOptional" : "0",
315
+ "dataType" : "Float16",
316
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
317
+ "shortDescription" : "",
318
+ "shape" : "[1, 1, 128]",
319
+ "name" : "sin",
320
+ "type" : "MultiArray"
321
+ },
322
+ {
323
+ "hasShapeFlexibility" : "0",
324
+ "isOptional" : "0",
325
+ "dataType" : "Float16",
326
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
327
+ "shortDescription" : "",
328
+ "shape" : "[1, 1, 1, 1024]",
329
+ "name" : "attention_mask",
330
+ "type" : "MultiArray"
331
+ },
332
+ {
333
+ "hasShapeFlexibility" : "0",
334
+ "isOptional" : "0",
335
+ "dataType" : "Int32",
336
+ "formattedType" : "MultiArray (Int32 1)",
337
+ "shortDescription" : "",
338
+ "shape" : "[1]",
339
+ "name" : "n_past",
340
+ "type" : "MultiArray"
341
+ }
342
+ ],
343
+ "defaultFunctionName" : "ar1",
344
+ "generatedClassName" : "qwen3_4B_part0_ar1_32",
345
+ "userDefinedMetadata" : {
346
+
347
+ },
348
+ "method" : "predict"
349
+ }
350
+ ]
qwen3-4B-part0-ar1-32.mlmodelc/model.mil ADDED
The diff for this file is too large to render. See raw diff
 
qwen3-4B-part0-ar1-32.mlmodelc/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eae57ae4cf993170b2df2099e487923d728c821cd1842ac23cba0b93990ecf81
3
+ size 807449408
qwen3-4B-part1-ar1-32.mlmodelc/analytics/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:846768a9d33810840860dbf42fd49f635d2f5ad3fdb34610e80e276241f7dd90
3
+ size 243
qwen3-4B-part1-ar1-32.mlmodelc/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6181f4eb9eec7d084dc9b7c68b9747a75634e25e74b39b9c3aa26dc526d2c5c
3
+ size 577
qwen3-4B-part1-ar1-32.mlmodelc/metadata.json ADDED
@@ -0,0 +1,350 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "metadataOutputVersion" : "3.0",
4
+ "storagePrecision" : "Float16",
5
+ "outputSchema" : [
6
+ {
7
+ "hasShapeFlexibility" : "0",
8
+ "isOptional" : "0",
9
+ "dataType" : "Float16",
10
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
11
+ "shortDescription" : "",
12
+ "shape" : "[1, 1, 2560]",
13
+ "name" : "output",
14
+ "type" : "MultiArray"
15
+ }
16
+ ],
17
+ "modelParameters" : [
18
+
19
+ ],
20
+ "specificationVersion" : 9,
21
+ "functions" : [
22
+ {
23
+ "inputSchema" : [
24
+ {
25
+ "hasShapeFlexibility" : "0",
26
+ "isOptional" : "0",
27
+ "dataType" : "Float16",
28
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
29
+ "shortDescription" : "",
30
+ "shape" : "[1, 1, 2560]",
31
+ "name" : "input_embedding",
32
+ "type" : "MultiArray"
33
+ },
34
+ {
35
+ "hasShapeFlexibility" : "0",
36
+ "isOptional" : "0",
37
+ "dataType" : "Float16",
38
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
39
+ "shortDescription" : "",
40
+ "shape" : "[1, 1, 128]",
41
+ "name" : "cos",
42
+ "type" : "MultiArray"
43
+ },
44
+ {
45
+ "hasShapeFlexibility" : "0",
46
+ "isOptional" : "0",
47
+ "dataType" : "Float16",
48
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
49
+ "shortDescription" : "",
50
+ "shape" : "[1, 1, 128]",
51
+ "name" : "sin",
52
+ "type" : "MultiArray"
53
+ },
54
+ {
55
+ "hasShapeFlexibility" : "0",
56
+ "isOptional" : "0",
57
+ "dataType" : "Float16",
58
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
59
+ "shortDescription" : "",
60
+ "shape" : "[1, 1, 1, 1024]",
61
+ "name" : "attention_mask",
62
+ "type" : "MultiArray"
63
+ },
64
+ {
65
+ "hasShapeFlexibility" : "0",
66
+ "isOptional" : "0",
67
+ "dataType" : "Int32",
68
+ "formattedType" : "MultiArray (Int32 1)",
69
+ "shortDescription" : "",
70
+ "shape" : "[1]",
71
+ "name" : "n_past",
72
+ "type" : "MultiArray"
73
+ }
74
+ ],
75
+ "computePrecision" : "Mixed (Float16, Int32)",
76
+ "storagePrecision" : "Float16",
77
+ "stateSchema" : [
78
+ {
79
+ "dataType" : "Float16",
80
+ "isOptional" : "0",
81
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
82
+ "shortDescription" : "",
83
+ "shape" : "[4, 8, 1024, 128]",
84
+ "name" : "key_cache",
85
+ "type" : "State"
86
+ },
87
+ {
88
+ "dataType" : "Float16",
89
+ "isOptional" : "0",
90
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
91
+ "shortDescription" : "",
92
+ "shape" : "[4, 8, 1024, 128]",
93
+ "name" : "value_cache",
94
+ "type" : "State"
95
+ }
96
+ ],
97
+ "outputSchema" : [
98
+ {
99
+ "hasShapeFlexibility" : "0",
100
+ "isOptional" : "0",
101
+ "dataType" : "Float16",
102
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
103
+ "shortDescription" : "",
104
+ "shape" : "[1, 1, 2560]",
105
+ "name" : "output",
106
+ "type" : "MultiArray"
107
+ }
108
+ ],
109
+ "name" : "ar1",
110
+ "mlProgramOperationTypeHistogram" : {
111
+ "Ios18.readState" : 10,
112
+ "Ios18.conv" : 28,
113
+ "Ios18.expandDims" : 26,
114
+ "Ios18.matmul" : 8,
115
+ "Ios18.concat" : 29,
116
+ "Ios18.add" : 21,
117
+ "Ios18.silu" : 4,
118
+ "Ios18.sliceUpdate" : 8,
119
+ "Ios18.sliceByIndex" : 40,
120
+ "Ios18.layerNorm" : 16,
121
+ "Ios18.transpose" : 36,
122
+ "Tile" : 8,
123
+ "Ios18.writeState" : 8,
124
+ "Ios18.reshape" : 32,
125
+ "Ios18.softmax" : 4,
126
+ "Ios18.squeeze" : 4,
127
+ "Ios18.mul" : 64
128
+ }
129
+ },
130
+ {
131
+ "inputSchema" : [
132
+ {
133
+ "hasShapeFlexibility" : "0",
134
+ "isOptional" : "0",
135
+ "dataType" : "Float16",
136
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
137
+ "shortDescription" : "",
138
+ "shape" : "[1, 32, 2560]",
139
+ "name" : "input_embedding",
140
+ "type" : "MultiArray"
141
+ },
142
+ {
143
+ "hasShapeFlexibility" : "0",
144
+ "isOptional" : "0",
145
+ "dataType" : "Float16",
146
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
147
+ "shortDescription" : "",
148
+ "shape" : "[1, 32, 128]",
149
+ "name" : "cos",
150
+ "type" : "MultiArray"
151
+ },
152
+ {
153
+ "hasShapeFlexibility" : "0",
154
+ "isOptional" : "0",
155
+ "dataType" : "Float16",
156
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
157
+ "shortDescription" : "",
158
+ "shape" : "[1, 32, 128]",
159
+ "name" : "sin",
160
+ "type" : "MultiArray"
161
+ },
162
+ {
163
+ "hasShapeFlexibility" : "0",
164
+ "isOptional" : "0",
165
+ "dataType" : "Float16",
166
+ "formattedType" : "MultiArray (Float16 1 × 1 × 32 × 1024)",
167
+ "shortDescription" : "",
168
+ "shape" : "[1, 1, 32, 1024]",
169
+ "name" : "attention_mask",
170
+ "type" : "MultiArray"
171
+ },
172
+ {
173
+ "hasShapeFlexibility" : "0",
174
+ "isOptional" : "0",
175
+ "dataType" : "Int32",
176
+ "formattedType" : "MultiArray (Int32 1)",
177
+ "shortDescription" : "",
178
+ "shape" : "[1]",
179
+ "name" : "n_past",
180
+ "type" : "MultiArray"
181
+ }
182
+ ],
183
+ "computePrecision" : "Mixed (Float16, Int32)",
184
+ "storagePrecision" : "Float16",
185
+ "stateSchema" : [
186
+ {
187
+ "dataType" : "Float16",
188
+ "isOptional" : "0",
189
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
190
+ "shortDescription" : "",
191
+ "shape" : "[4, 8, 1024, 128]",
192
+ "name" : "key_cache",
193
+ "type" : "State"
194
+ },
195
+ {
196
+ "dataType" : "Float16",
197
+ "isOptional" : "0",
198
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
199
+ "shortDescription" : "",
200
+ "shape" : "[4, 8, 1024, 128]",
201
+ "name" : "value_cache",
202
+ "type" : "State"
203
+ }
204
+ ],
205
+ "outputSchema" : [
206
+ {
207
+ "hasShapeFlexibility" : "0",
208
+ "isOptional" : "0",
209
+ "dataType" : "Float16",
210
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
211
+ "shortDescription" : "",
212
+ "shape" : "[1, 32, 2560]",
213
+ "name" : "output",
214
+ "type" : "MultiArray"
215
+ }
216
+ ],
217
+ "name" : "ar32",
218
+ "mlProgramOperationTypeHistogram" : {
219
+ "Ios18.readState" : 10,
220
+ "Ios18.conv" : 28,
221
+ "Ios18.expandDims" : 26,
222
+ "Ios18.matmul" : 8,
223
+ "Ios18.concat" : 29,
224
+ "Ios18.add" : 21,
225
+ "Ios18.silu" : 4,
226
+ "Ios18.sliceUpdate" : 8,
227
+ "Ios18.sliceByIndex" : 40,
228
+ "Ios18.layerNorm" : 16,
229
+ "Ios18.transpose" : 36,
230
+ "Tile" : 8,
231
+ "Ios18.writeState" : 8,
232
+ "Ios18.reshape" : 32,
233
+ "Ios18.softmax" : 4,
234
+ "Ios18.squeeze" : 4,
235
+ "Ios18.mul" : 64
236
+ }
237
+ }
238
+ ],
239
+ "mlProgramOperationTypeHistogram" : {
240
+ "Ios18.readState" : 10,
241
+ "Ios18.conv" : 28,
242
+ "Ios18.expandDims" : 26,
243
+ "Ios18.matmul" : 8,
244
+ "Ios18.concat" : 29,
245
+ "Ios18.add" : 21,
246
+ "Ios18.silu" : 4,
247
+ "Ios18.sliceUpdate" : 8,
248
+ "Ios18.sliceByIndex" : 40,
249
+ "Ios18.layerNorm" : 16,
250
+ "Ios18.transpose" : 36,
251
+ "Tile" : 8,
252
+ "Ios18.writeState" : 8,
253
+ "Ios18.reshape" : 32,
254
+ "Ios18.softmax" : 4,
255
+ "Ios18.squeeze" : 4,
256
+ "Ios18.mul" : 64
257
+ },
258
+ "isUpdatable" : "0",
259
+ "stateSchema" : [
260
+ {
261
+ "dataType" : "Float16",
262
+ "isOptional" : "0",
263
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
264
+ "shortDescription" : "",
265
+ "shape" : "[4, 8, 1024, 128]",
266
+ "name" : "key_cache",
267
+ "type" : "State"
268
+ },
269
+ {
270
+ "dataType" : "Float16",
271
+ "isOptional" : "0",
272
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
273
+ "shortDescription" : "",
274
+ "shape" : "[4, 8, 1024, 128]",
275
+ "name" : "value_cache",
276
+ "type" : "State"
277
+ }
278
+ ],
279
+ "availability" : {
280
+ "macOS" : "15.0",
281
+ "tvOS" : "18.0",
282
+ "visionOS" : "2.0",
283
+ "watchOS" : "11.0",
284
+ "iOS" : "18.0",
285
+ "macCatalyst" : "18.0"
286
+ },
287
+ "computePrecision" : "Mixed (Float16, Int32)",
288
+ "modelType" : {
289
+ "name" : "MLModelType_mlProgram"
290
+ },
291
+ "inputSchema" : [
292
+ {
293
+ "hasShapeFlexibility" : "0",
294
+ "isOptional" : "0",
295
+ "dataType" : "Float16",
296
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
297
+ "shortDescription" : "",
298
+ "shape" : "[1, 1, 2560]",
299
+ "name" : "input_embedding",
300
+ "type" : "MultiArray"
301
+ },
302
+ {
303
+ "hasShapeFlexibility" : "0",
304
+ "isOptional" : "0",
305
+ "dataType" : "Float16",
306
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
307
+ "shortDescription" : "",
308
+ "shape" : "[1, 1, 128]",
309
+ "name" : "cos",
310
+ "type" : "MultiArray"
311
+ },
312
+ {
313
+ "hasShapeFlexibility" : "0",
314
+ "isOptional" : "0",
315
+ "dataType" : "Float16",
316
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
317
+ "shortDescription" : "",
318
+ "shape" : "[1, 1, 128]",
319
+ "name" : "sin",
320
+ "type" : "MultiArray"
321
+ },
322
+ {
323
+ "hasShapeFlexibility" : "0",
324
+ "isOptional" : "0",
325
+ "dataType" : "Float16",
326
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
327
+ "shortDescription" : "",
328
+ "shape" : "[1, 1, 1, 1024]",
329
+ "name" : "attention_mask",
330
+ "type" : "MultiArray"
331
+ },
332
+ {
333
+ "hasShapeFlexibility" : "0",
334
+ "isOptional" : "0",
335
+ "dataType" : "Int32",
336
+ "formattedType" : "MultiArray (Int32 1)",
337
+ "shortDescription" : "",
338
+ "shape" : "[1]",
339
+ "name" : "n_past",
340
+ "type" : "MultiArray"
341
+ }
342
+ ],
343
+ "defaultFunctionName" : "ar1",
344
+ "generatedClassName" : "qwen3_4B_part1_ar1_32",
345
+ "userDefinedMetadata" : {
346
+
347
+ },
348
+ "method" : "predict"
349
+ }
350
+ ]
qwen3-4B-part1-ar1-32.mlmodelc/model.mil ADDED
The diff for this file is too large to render. See raw diff
 
qwen3-4B-part1-ar1-32.mlmodelc/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6907cc79cdcd6b3d5116595a6dc4eaf5bffd2870e21afb99d656953cc6621ee8
3
+ size 807449408
qwen3-4B-part2-ar1-32.mlmodelc/analytics/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:846768a9d33810840860dbf42fd49f635d2f5ad3fdb34610e80e276241f7dd90
3
+ size 243
qwen3-4B-part2-ar1-32.mlmodelc/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6181f4eb9eec7d084dc9b7c68b9747a75634e25e74b39b9c3aa26dc526d2c5c
3
+ size 577
qwen3-4B-part2-ar1-32.mlmodelc/metadata.json ADDED
@@ -0,0 +1,350 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "metadataOutputVersion" : "3.0",
4
+ "storagePrecision" : "Float16",
5
+ "outputSchema" : [
6
+ {
7
+ "hasShapeFlexibility" : "0",
8
+ "isOptional" : "0",
9
+ "dataType" : "Float16",
10
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
11
+ "shortDescription" : "",
12
+ "shape" : "[1, 1, 2560]",
13
+ "name" : "output",
14
+ "type" : "MultiArray"
15
+ }
16
+ ],
17
+ "modelParameters" : [
18
+
19
+ ],
20
+ "specificationVersion" : 9,
21
+ "functions" : [
22
+ {
23
+ "inputSchema" : [
24
+ {
25
+ "hasShapeFlexibility" : "0",
26
+ "isOptional" : "0",
27
+ "dataType" : "Float16",
28
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
29
+ "shortDescription" : "",
30
+ "shape" : "[1, 1, 2560]",
31
+ "name" : "input_embedding",
32
+ "type" : "MultiArray"
33
+ },
34
+ {
35
+ "hasShapeFlexibility" : "0",
36
+ "isOptional" : "0",
37
+ "dataType" : "Float16",
38
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
39
+ "shortDescription" : "",
40
+ "shape" : "[1, 1, 128]",
41
+ "name" : "cos",
42
+ "type" : "MultiArray"
43
+ },
44
+ {
45
+ "hasShapeFlexibility" : "0",
46
+ "isOptional" : "0",
47
+ "dataType" : "Float16",
48
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
49
+ "shortDescription" : "",
50
+ "shape" : "[1, 1, 128]",
51
+ "name" : "sin",
52
+ "type" : "MultiArray"
53
+ },
54
+ {
55
+ "hasShapeFlexibility" : "0",
56
+ "isOptional" : "0",
57
+ "dataType" : "Float16",
58
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
59
+ "shortDescription" : "",
60
+ "shape" : "[1, 1, 1, 1024]",
61
+ "name" : "attention_mask",
62
+ "type" : "MultiArray"
63
+ },
64
+ {
65
+ "hasShapeFlexibility" : "0",
66
+ "isOptional" : "0",
67
+ "dataType" : "Int32",
68
+ "formattedType" : "MultiArray (Int32 1)",
69
+ "shortDescription" : "",
70
+ "shape" : "[1]",
71
+ "name" : "n_past",
72
+ "type" : "MultiArray"
73
+ }
74
+ ],
75
+ "computePrecision" : "Mixed (Float16, Int32)",
76
+ "storagePrecision" : "Float16",
77
+ "stateSchema" : [
78
+ {
79
+ "dataType" : "Float16",
80
+ "isOptional" : "0",
81
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
82
+ "shortDescription" : "",
83
+ "shape" : "[4, 8, 1024, 128]",
84
+ "name" : "key_cache",
85
+ "type" : "State"
86
+ },
87
+ {
88
+ "dataType" : "Float16",
89
+ "isOptional" : "0",
90
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
91
+ "shortDescription" : "",
92
+ "shape" : "[4, 8, 1024, 128]",
93
+ "name" : "value_cache",
94
+ "type" : "State"
95
+ }
96
+ ],
97
+ "outputSchema" : [
98
+ {
99
+ "hasShapeFlexibility" : "0",
100
+ "isOptional" : "0",
101
+ "dataType" : "Float16",
102
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
103
+ "shortDescription" : "",
104
+ "shape" : "[1, 1, 2560]",
105
+ "name" : "output",
106
+ "type" : "MultiArray"
107
+ }
108
+ ],
109
+ "name" : "ar1",
110
+ "mlProgramOperationTypeHistogram" : {
111
+ "Ios18.readState" : 10,
112
+ "Ios18.conv" : 28,
113
+ "Ios18.expandDims" : 26,
114
+ "Ios18.matmul" : 8,
115
+ "Ios18.concat" : 29,
116
+ "Ios18.add" : 21,
117
+ "Ios18.silu" : 4,
118
+ "Ios18.sliceUpdate" : 8,
119
+ "Ios18.sliceByIndex" : 40,
120
+ "Ios18.layerNorm" : 16,
121
+ "Ios18.transpose" : 36,
122
+ "Tile" : 8,
123
+ "Ios18.writeState" : 8,
124
+ "Ios18.reshape" : 32,
125
+ "Ios18.softmax" : 4,
126
+ "Ios18.squeeze" : 4,
127
+ "Ios18.mul" : 64
128
+ }
129
+ },
130
+ {
131
+ "inputSchema" : [
132
+ {
133
+ "hasShapeFlexibility" : "0",
134
+ "isOptional" : "0",
135
+ "dataType" : "Float16",
136
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
137
+ "shortDescription" : "",
138
+ "shape" : "[1, 32, 2560]",
139
+ "name" : "input_embedding",
140
+ "type" : "MultiArray"
141
+ },
142
+ {
143
+ "hasShapeFlexibility" : "0",
144
+ "isOptional" : "0",
145
+ "dataType" : "Float16",
146
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
147
+ "shortDescription" : "",
148
+ "shape" : "[1, 32, 128]",
149
+ "name" : "cos",
150
+ "type" : "MultiArray"
151
+ },
152
+ {
153
+ "hasShapeFlexibility" : "0",
154
+ "isOptional" : "0",
155
+ "dataType" : "Float16",
156
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
157
+ "shortDescription" : "",
158
+ "shape" : "[1, 32, 128]",
159
+ "name" : "sin",
160
+ "type" : "MultiArray"
161
+ },
162
+ {
163
+ "hasShapeFlexibility" : "0",
164
+ "isOptional" : "0",
165
+ "dataType" : "Float16",
166
+ "formattedType" : "MultiArray (Float16 1 × 1 × 32 × 1024)",
167
+ "shortDescription" : "",
168
+ "shape" : "[1, 1, 32, 1024]",
169
+ "name" : "attention_mask",
170
+ "type" : "MultiArray"
171
+ },
172
+ {
173
+ "hasShapeFlexibility" : "0",
174
+ "isOptional" : "0",
175
+ "dataType" : "Int32",
176
+ "formattedType" : "MultiArray (Int32 1)",
177
+ "shortDescription" : "",
178
+ "shape" : "[1]",
179
+ "name" : "n_past",
180
+ "type" : "MultiArray"
181
+ }
182
+ ],
183
+ "computePrecision" : "Mixed (Float16, Int32)",
184
+ "storagePrecision" : "Float16",
185
+ "stateSchema" : [
186
+ {
187
+ "dataType" : "Float16",
188
+ "isOptional" : "0",
189
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
190
+ "shortDescription" : "",
191
+ "shape" : "[4, 8, 1024, 128]",
192
+ "name" : "key_cache",
193
+ "type" : "State"
194
+ },
195
+ {
196
+ "dataType" : "Float16",
197
+ "isOptional" : "0",
198
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
199
+ "shortDescription" : "",
200
+ "shape" : "[4, 8, 1024, 128]",
201
+ "name" : "value_cache",
202
+ "type" : "State"
203
+ }
204
+ ],
205
+ "outputSchema" : [
206
+ {
207
+ "hasShapeFlexibility" : "0",
208
+ "isOptional" : "0",
209
+ "dataType" : "Float16",
210
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
211
+ "shortDescription" : "",
212
+ "shape" : "[1, 32, 2560]",
213
+ "name" : "output",
214
+ "type" : "MultiArray"
215
+ }
216
+ ],
217
+ "name" : "ar32",
218
+ "mlProgramOperationTypeHistogram" : {
219
+ "Ios18.readState" : 10,
220
+ "Ios18.conv" : 28,
221
+ "Ios18.expandDims" : 26,
222
+ "Ios18.matmul" : 8,
223
+ "Ios18.concat" : 29,
224
+ "Ios18.add" : 21,
225
+ "Ios18.silu" : 4,
226
+ "Ios18.sliceUpdate" : 8,
227
+ "Ios18.sliceByIndex" : 40,
228
+ "Ios18.layerNorm" : 16,
229
+ "Ios18.transpose" : 36,
230
+ "Tile" : 8,
231
+ "Ios18.writeState" : 8,
232
+ "Ios18.reshape" : 32,
233
+ "Ios18.softmax" : 4,
234
+ "Ios18.squeeze" : 4,
235
+ "Ios18.mul" : 64
236
+ }
237
+ }
238
+ ],
239
+ "mlProgramOperationTypeHistogram" : {
240
+ "Ios18.readState" : 10,
241
+ "Ios18.conv" : 28,
242
+ "Ios18.expandDims" : 26,
243
+ "Ios18.matmul" : 8,
244
+ "Ios18.concat" : 29,
245
+ "Ios18.add" : 21,
246
+ "Ios18.silu" : 4,
247
+ "Ios18.sliceUpdate" : 8,
248
+ "Ios18.sliceByIndex" : 40,
249
+ "Ios18.layerNorm" : 16,
250
+ "Ios18.transpose" : 36,
251
+ "Tile" : 8,
252
+ "Ios18.writeState" : 8,
253
+ "Ios18.reshape" : 32,
254
+ "Ios18.softmax" : 4,
255
+ "Ios18.squeeze" : 4,
256
+ "Ios18.mul" : 64
257
+ },
258
+ "isUpdatable" : "0",
259
+ "stateSchema" : [
260
+ {
261
+ "dataType" : "Float16",
262
+ "isOptional" : "0",
263
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
264
+ "shortDescription" : "",
265
+ "shape" : "[4, 8, 1024, 128]",
266
+ "name" : "key_cache",
267
+ "type" : "State"
268
+ },
269
+ {
270
+ "dataType" : "Float16",
271
+ "isOptional" : "0",
272
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
273
+ "shortDescription" : "",
274
+ "shape" : "[4, 8, 1024, 128]",
275
+ "name" : "value_cache",
276
+ "type" : "State"
277
+ }
278
+ ],
279
+ "availability" : {
280
+ "macOS" : "15.0",
281
+ "tvOS" : "18.0",
282
+ "visionOS" : "2.0",
283
+ "watchOS" : "11.0",
284
+ "iOS" : "18.0",
285
+ "macCatalyst" : "18.0"
286
+ },
287
+ "computePrecision" : "Mixed (Float16, Int32)",
288
+ "modelType" : {
289
+ "name" : "MLModelType_mlProgram"
290
+ },
291
+ "inputSchema" : [
292
+ {
293
+ "hasShapeFlexibility" : "0",
294
+ "isOptional" : "0",
295
+ "dataType" : "Float16",
296
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
297
+ "shortDescription" : "",
298
+ "shape" : "[1, 1, 2560]",
299
+ "name" : "input_embedding",
300
+ "type" : "MultiArray"
301
+ },
302
+ {
303
+ "hasShapeFlexibility" : "0",
304
+ "isOptional" : "0",
305
+ "dataType" : "Float16",
306
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
307
+ "shortDescription" : "",
308
+ "shape" : "[1, 1, 128]",
309
+ "name" : "cos",
310
+ "type" : "MultiArray"
311
+ },
312
+ {
313
+ "hasShapeFlexibility" : "0",
314
+ "isOptional" : "0",
315
+ "dataType" : "Float16",
316
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
317
+ "shortDescription" : "",
318
+ "shape" : "[1, 1, 128]",
319
+ "name" : "sin",
320
+ "type" : "MultiArray"
321
+ },
322
+ {
323
+ "hasShapeFlexibility" : "0",
324
+ "isOptional" : "0",
325
+ "dataType" : "Float16",
326
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
327
+ "shortDescription" : "",
328
+ "shape" : "[1, 1, 1, 1024]",
329
+ "name" : "attention_mask",
330
+ "type" : "MultiArray"
331
+ },
332
+ {
333
+ "hasShapeFlexibility" : "0",
334
+ "isOptional" : "0",
335
+ "dataType" : "Int32",
336
+ "formattedType" : "MultiArray (Int32 1)",
337
+ "shortDescription" : "",
338
+ "shape" : "[1]",
339
+ "name" : "n_past",
340
+ "type" : "MultiArray"
341
+ }
342
+ ],
343
+ "defaultFunctionName" : "ar1",
344
+ "generatedClassName" : "qwen3_4B_part2_ar1_32",
345
+ "userDefinedMetadata" : {
346
+
347
+ },
348
+ "method" : "predict"
349
+ }
350
+ ]
qwen3-4B-part2-ar1-32.mlmodelc/model.mil ADDED
The diff for this file is too large to render. See raw diff
 
qwen3-4B-part2-ar1-32.mlmodelc/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5cf060b18d98fa01eafd896c5dba229bda8857263d7582f210d0ef09adb84800
3
+ size 807449408
qwen3-4B-part3-ar1-32.mlmodelc/analytics/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:846768a9d33810840860dbf42fd49f635d2f5ad3fdb34610e80e276241f7dd90
3
+ size 243
qwen3-4B-part3-ar1-32.mlmodelc/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6181f4eb9eec7d084dc9b7c68b9747a75634e25e74b39b9c3aa26dc526d2c5c
3
+ size 577
qwen3-4B-part3-ar1-32.mlmodelc/metadata.json ADDED
@@ -0,0 +1,350 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "metadataOutputVersion" : "3.0",
4
+ "storagePrecision" : "Float16",
5
+ "outputSchema" : [
6
+ {
7
+ "hasShapeFlexibility" : "0",
8
+ "isOptional" : "0",
9
+ "dataType" : "Float16",
10
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
11
+ "shortDescription" : "",
12
+ "shape" : "[1, 1, 2560]",
13
+ "name" : "output",
14
+ "type" : "MultiArray"
15
+ }
16
+ ],
17
+ "modelParameters" : [
18
+
19
+ ],
20
+ "specificationVersion" : 9,
21
+ "functions" : [
22
+ {
23
+ "inputSchema" : [
24
+ {
25
+ "hasShapeFlexibility" : "0",
26
+ "isOptional" : "0",
27
+ "dataType" : "Float16",
28
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
29
+ "shortDescription" : "",
30
+ "shape" : "[1, 1, 2560]",
31
+ "name" : "input_embedding",
32
+ "type" : "MultiArray"
33
+ },
34
+ {
35
+ "hasShapeFlexibility" : "0",
36
+ "isOptional" : "0",
37
+ "dataType" : "Float16",
38
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
39
+ "shortDescription" : "",
40
+ "shape" : "[1, 1, 128]",
41
+ "name" : "cos",
42
+ "type" : "MultiArray"
43
+ },
44
+ {
45
+ "hasShapeFlexibility" : "0",
46
+ "isOptional" : "0",
47
+ "dataType" : "Float16",
48
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
49
+ "shortDescription" : "",
50
+ "shape" : "[1, 1, 128]",
51
+ "name" : "sin",
52
+ "type" : "MultiArray"
53
+ },
54
+ {
55
+ "hasShapeFlexibility" : "0",
56
+ "isOptional" : "0",
57
+ "dataType" : "Float16",
58
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
59
+ "shortDescription" : "",
60
+ "shape" : "[1, 1, 1, 1024]",
61
+ "name" : "attention_mask",
62
+ "type" : "MultiArray"
63
+ },
64
+ {
65
+ "hasShapeFlexibility" : "0",
66
+ "isOptional" : "0",
67
+ "dataType" : "Int32",
68
+ "formattedType" : "MultiArray (Int32 1)",
69
+ "shortDescription" : "",
70
+ "shape" : "[1]",
71
+ "name" : "n_past",
72
+ "type" : "MultiArray"
73
+ }
74
+ ],
75
+ "computePrecision" : "Mixed (Float16, Int32)",
76
+ "storagePrecision" : "Float16",
77
+ "stateSchema" : [
78
+ {
79
+ "dataType" : "Float16",
80
+ "isOptional" : "0",
81
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
82
+ "shortDescription" : "",
83
+ "shape" : "[4, 8, 1024, 128]",
84
+ "name" : "key_cache",
85
+ "type" : "State"
86
+ },
87
+ {
88
+ "dataType" : "Float16",
89
+ "isOptional" : "0",
90
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
91
+ "shortDescription" : "",
92
+ "shape" : "[4, 8, 1024, 128]",
93
+ "name" : "value_cache",
94
+ "type" : "State"
95
+ }
96
+ ],
97
+ "outputSchema" : [
98
+ {
99
+ "hasShapeFlexibility" : "0",
100
+ "isOptional" : "0",
101
+ "dataType" : "Float16",
102
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
103
+ "shortDescription" : "",
104
+ "shape" : "[1, 1, 2560]",
105
+ "name" : "output",
106
+ "type" : "MultiArray"
107
+ }
108
+ ],
109
+ "name" : "ar1",
110
+ "mlProgramOperationTypeHistogram" : {
111
+ "Ios18.readState" : 10,
112
+ "Ios18.conv" : 28,
113
+ "Ios18.expandDims" : 26,
114
+ "Ios18.matmul" : 8,
115
+ "Ios18.concat" : 29,
116
+ "Ios18.add" : 21,
117
+ "Ios18.silu" : 4,
118
+ "Ios18.sliceUpdate" : 8,
119
+ "Ios18.sliceByIndex" : 40,
120
+ "Ios18.layerNorm" : 16,
121
+ "Ios18.transpose" : 36,
122
+ "Tile" : 8,
123
+ "Ios18.writeState" : 8,
124
+ "Ios18.reshape" : 32,
125
+ "Ios18.softmax" : 4,
126
+ "Ios18.squeeze" : 4,
127
+ "Ios18.mul" : 64
128
+ }
129
+ },
130
+ {
131
+ "inputSchema" : [
132
+ {
133
+ "hasShapeFlexibility" : "0",
134
+ "isOptional" : "0",
135
+ "dataType" : "Float16",
136
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
137
+ "shortDescription" : "",
138
+ "shape" : "[1, 32, 2560]",
139
+ "name" : "input_embedding",
140
+ "type" : "MultiArray"
141
+ },
142
+ {
143
+ "hasShapeFlexibility" : "0",
144
+ "isOptional" : "0",
145
+ "dataType" : "Float16",
146
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
147
+ "shortDescription" : "",
148
+ "shape" : "[1, 32, 128]",
149
+ "name" : "cos",
150
+ "type" : "MultiArray"
151
+ },
152
+ {
153
+ "hasShapeFlexibility" : "0",
154
+ "isOptional" : "0",
155
+ "dataType" : "Float16",
156
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
157
+ "shortDescription" : "",
158
+ "shape" : "[1, 32, 128]",
159
+ "name" : "sin",
160
+ "type" : "MultiArray"
161
+ },
162
+ {
163
+ "hasShapeFlexibility" : "0",
164
+ "isOptional" : "0",
165
+ "dataType" : "Float16",
166
+ "formattedType" : "MultiArray (Float16 1 × 1 × 32 × 1024)",
167
+ "shortDescription" : "",
168
+ "shape" : "[1, 1, 32, 1024]",
169
+ "name" : "attention_mask",
170
+ "type" : "MultiArray"
171
+ },
172
+ {
173
+ "hasShapeFlexibility" : "0",
174
+ "isOptional" : "0",
175
+ "dataType" : "Int32",
176
+ "formattedType" : "MultiArray (Int32 1)",
177
+ "shortDescription" : "",
178
+ "shape" : "[1]",
179
+ "name" : "n_past",
180
+ "type" : "MultiArray"
181
+ }
182
+ ],
183
+ "computePrecision" : "Mixed (Float16, Int32)",
184
+ "storagePrecision" : "Float16",
185
+ "stateSchema" : [
186
+ {
187
+ "dataType" : "Float16",
188
+ "isOptional" : "0",
189
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
190
+ "shortDescription" : "",
191
+ "shape" : "[4, 8, 1024, 128]",
192
+ "name" : "key_cache",
193
+ "type" : "State"
194
+ },
195
+ {
196
+ "dataType" : "Float16",
197
+ "isOptional" : "0",
198
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
199
+ "shortDescription" : "",
200
+ "shape" : "[4, 8, 1024, 128]",
201
+ "name" : "value_cache",
202
+ "type" : "State"
203
+ }
204
+ ],
205
+ "outputSchema" : [
206
+ {
207
+ "hasShapeFlexibility" : "0",
208
+ "isOptional" : "0",
209
+ "dataType" : "Float16",
210
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
211
+ "shortDescription" : "",
212
+ "shape" : "[1, 32, 2560]",
213
+ "name" : "output",
214
+ "type" : "MultiArray"
215
+ }
216
+ ],
217
+ "name" : "ar32",
218
+ "mlProgramOperationTypeHistogram" : {
219
+ "Ios18.readState" : 10,
220
+ "Ios18.conv" : 28,
221
+ "Ios18.expandDims" : 26,
222
+ "Ios18.matmul" : 8,
223
+ "Ios18.concat" : 29,
224
+ "Ios18.add" : 21,
225
+ "Ios18.silu" : 4,
226
+ "Ios18.sliceUpdate" : 8,
227
+ "Ios18.sliceByIndex" : 40,
228
+ "Ios18.layerNorm" : 16,
229
+ "Ios18.transpose" : 36,
230
+ "Tile" : 8,
231
+ "Ios18.writeState" : 8,
232
+ "Ios18.reshape" : 32,
233
+ "Ios18.softmax" : 4,
234
+ "Ios18.squeeze" : 4,
235
+ "Ios18.mul" : 64
236
+ }
237
+ }
238
+ ],
239
+ "mlProgramOperationTypeHistogram" : {
240
+ "Ios18.readState" : 10,
241
+ "Ios18.conv" : 28,
242
+ "Ios18.expandDims" : 26,
243
+ "Ios18.matmul" : 8,
244
+ "Ios18.concat" : 29,
245
+ "Ios18.add" : 21,
246
+ "Ios18.silu" : 4,
247
+ "Ios18.sliceUpdate" : 8,
248
+ "Ios18.sliceByIndex" : 40,
249
+ "Ios18.layerNorm" : 16,
250
+ "Ios18.transpose" : 36,
251
+ "Tile" : 8,
252
+ "Ios18.writeState" : 8,
253
+ "Ios18.reshape" : 32,
254
+ "Ios18.softmax" : 4,
255
+ "Ios18.squeeze" : 4,
256
+ "Ios18.mul" : 64
257
+ },
258
+ "isUpdatable" : "0",
259
+ "stateSchema" : [
260
+ {
261
+ "dataType" : "Float16",
262
+ "isOptional" : "0",
263
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
264
+ "shortDescription" : "",
265
+ "shape" : "[4, 8, 1024, 128]",
266
+ "name" : "key_cache",
267
+ "type" : "State"
268
+ },
269
+ {
270
+ "dataType" : "Float16",
271
+ "isOptional" : "0",
272
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
273
+ "shortDescription" : "",
274
+ "shape" : "[4, 8, 1024, 128]",
275
+ "name" : "value_cache",
276
+ "type" : "State"
277
+ }
278
+ ],
279
+ "availability" : {
280
+ "macOS" : "15.0",
281
+ "tvOS" : "18.0",
282
+ "visionOS" : "2.0",
283
+ "watchOS" : "11.0",
284
+ "iOS" : "18.0",
285
+ "macCatalyst" : "18.0"
286
+ },
287
+ "computePrecision" : "Mixed (Float16, Int32)",
288
+ "modelType" : {
289
+ "name" : "MLModelType_mlProgram"
290
+ },
291
+ "inputSchema" : [
292
+ {
293
+ "hasShapeFlexibility" : "0",
294
+ "isOptional" : "0",
295
+ "dataType" : "Float16",
296
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
297
+ "shortDescription" : "",
298
+ "shape" : "[1, 1, 2560]",
299
+ "name" : "input_embedding",
300
+ "type" : "MultiArray"
301
+ },
302
+ {
303
+ "hasShapeFlexibility" : "0",
304
+ "isOptional" : "0",
305
+ "dataType" : "Float16",
306
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
307
+ "shortDescription" : "",
308
+ "shape" : "[1, 1, 128]",
309
+ "name" : "cos",
310
+ "type" : "MultiArray"
311
+ },
312
+ {
313
+ "hasShapeFlexibility" : "0",
314
+ "isOptional" : "0",
315
+ "dataType" : "Float16",
316
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
317
+ "shortDescription" : "",
318
+ "shape" : "[1, 1, 128]",
319
+ "name" : "sin",
320
+ "type" : "MultiArray"
321
+ },
322
+ {
323
+ "hasShapeFlexibility" : "0",
324
+ "isOptional" : "0",
325
+ "dataType" : "Float16",
326
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
327
+ "shortDescription" : "",
328
+ "shape" : "[1, 1, 1, 1024]",
329
+ "name" : "attention_mask",
330
+ "type" : "MultiArray"
331
+ },
332
+ {
333
+ "hasShapeFlexibility" : "0",
334
+ "isOptional" : "0",
335
+ "dataType" : "Int32",
336
+ "formattedType" : "MultiArray (Int32 1)",
337
+ "shortDescription" : "",
338
+ "shape" : "[1]",
339
+ "name" : "n_past",
340
+ "type" : "MultiArray"
341
+ }
342
+ ],
343
+ "defaultFunctionName" : "ar1",
344
+ "generatedClassName" : "qwen3_4B_part3_ar1_32",
345
+ "userDefinedMetadata" : {
346
+
347
+ },
348
+ "method" : "predict"
349
+ }
350
+ ]
qwen3-4B-part3-ar1-32.mlmodelc/model.mil ADDED
The diff for this file is too large to render. See raw diff
 
qwen3-4B-part3-ar1-32.mlmodelc/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c88d073b3a813aa25549bd2b9a615e367bf5936e6115b0151dde43d6aa2959dc
3
+ size 807449408
qwen3-4B-part4-ar1-32.mlmodelc/analytics/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:846768a9d33810840860dbf42fd49f635d2f5ad3fdb34610e80e276241f7dd90
3
+ size 243
qwen3-4B-part4-ar1-32.mlmodelc/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6181f4eb9eec7d084dc9b7c68b9747a75634e25e74b39b9c3aa26dc526d2c5c
3
+ size 577
qwen3-4B-part4-ar1-32.mlmodelc/metadata.json ADDED
@@ -0,0 +1,350 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "metadataOutputVersion" : "3.0",
4
+ "storagePrecision" : "Float16",
5
+ "outputSchema" : [
6
+ {
7
+ "hasShapeFlexibility" : "0",
8
+ "isOptional" : "0",
9
+ "dataType" : "Float16",
10
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
11
+ "shortDescription" : "",
12
+ "shape" : "[1, 1, 2560]",
13
+ "name" : "output",
14
+ "type" : "MultiArray"
15
+ }
16
+ ],
17
+ "modelParameters" : [
18
+
19
+ ],
20
+ "specificationVersion" : 9,
21
+ "functions" : [
22
+ {
23
+ "inputSchema" : [
24
+ {
25
+ "hasShapeFlexibility" : "0",
26
+ "isOptional" : "0",
27
+ "dataType" : "Float16",
28
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
29
+ "shortDescription" : "",
30
+ "shape" : "[1, 1, 2560]",
31
+ "name" : "input_embedding",
32
+ "type" : "MultiArray"
33
+ },
34
+ {
35
+ "hasShapeFlexibility" : "0",
36
+ "isOptional" : "0",
37
+ "dataType" : "Float16",
38
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
39
+ "shortDescription" : "",
40
+ "shape" : "[1, 1, 128]",
41
+ "name" : "cos",
42
+ "type" : "MultiArray"
43
+ },
44
+ {
45
+ "hasShapeFlexibility" : "0",
46
+ "isOptional" : "0",
47
+ "dataType" : "Float16",
48
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
49
+ "shortDescription" : "",
50
+ "shape" : "[1, 1, 128]",
51
+ "name" : "sin",
52
+ "type" : "MultiArray"
53
+ },
54
+ {
55
+ "hasShapeFlexibility" : "0",
56
+ "isOptional" : "0",
57
+ "dataType" : "Float16",
58
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
59
+ "shortDescription" : "",
60
+ "shape" : "[1, 1, 1, 1024]",
61
+ "name" : "attention_mask",
62
+ "type" : "MultiArray"
63
+ },
64
+ {
65
+ "hasShapeFlexibility" : "0",
66
+ "isOptional" : "0",
67
+ "dataType" : "Int32",
68
+ "formattedType" : "MultiArray (Int32 1)",
69
+ "shortDescription" : "",
70
+ "shape" : "[1]",
71
+ "name" : "n_past",
72
+ "type" : "MultiArray"
73
+ }
74
+ ],
75
+ "computePrecision" : "Mixed (Float16, Int32)",
76
+ "storagePrecision" : "Float16",
77
+ "stateSchema" : [
78
+ {
79
+ "dataType" : "Float16",
80
+ "isOptional" : "0",
81
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
82
+ "shortDescription" : "",
83
+ "shape" : "[4, 8, 1024, 128]",
84
+ "name" : "key_cache",
85
+ "type" : "State"
86
+ },
87
+ {
88
+ "dataType" : "Float16",
89
+ "isOptional" : "0",
90
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
91
+ "shortDescription" : "",
92
+ "shape" : "[4, 8, 1024, 128]",
93
+ "name" : "value_cache",
94
+ "type" : "State"
95
+ }
96
+ ],
97
+ "outputSchema" : [
98
+ {
99
+ "hasShapeFlexibility" : "0",
100
+ "isOptional" : "0",
101
+ "dataType" : "Float16",
102
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
103
+ "shortDescription" : "",
104
+ "shape" : "[1, 1, 2560]",
105
+ "name" : "output",
106
+ "type" : "MultiArray"
107
+ }
108
+ ],
109
+ "name" : "ar1",
110
+ "mlProgramOperationTypeHistogram" : {
111
+ "Ios18.readState" : 10,
112
+ "Ios18.conv" : 28,
113
+ "Ios18.expandDims" : 26,
114
+ "Ios18.matmul" : 8,
115
+ "Ios18.concat" : 29,
116
+ "Ios18.add" : 21,
117
+ "Ios18.silu" : 4,
118
+ "Ios18.sliceUpdate" : 8,
119
+ "Ios18.sliceByIndex" : 40,
120
+ "Ios18.layerNorm" : 16,
121
+ "Ios18.transpose" : 36,
122
+ "Tile" : 8,
123
+ "Ios18.writeState" : 8,
124
+ "Ios18.reshape" : 32,
125
+ "Ios18.softmax" : 4,
126
+ "Ios18.squeeze" : 4,
127
+ "Ios18.mul" : 64
128
+ }
129
+ },
130
+ {
131
+ "inputSchema" : [
132
+ {
133
+ "hasShapeFlexibility" : "0",
134
+ "isOptional" : "0",
135
+ "dataType" : "Float16",
136
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
137
+ "shortDescription" : "",
138
+ "shape" : "[1, 32, 2560]",
139
+ "name" : "input_embedding",
140
+ "type" : "MultiArray"
141
+ },
142
+ {
143
+ "hasShapeFlexibility" : "0",
144
+ "isOptional" : "0",
145
+ "dataType" : "Float16",
146
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
147
+ "shortDescription" : "",
148
+ "shape" : "[1, 32, 128]",
149
+ "name" : "cos",
150
+ "type" : "MultiArray"
151
+ },
152
+ {
153
+ "hasShapeFlexibility" : "0",
154
+ "isOptional" : "0",
155
+ "dataType" : "Float16",
156
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
157
+ "shortDescription" : "",
158
+ "shape" : "[1, 32, 128]",
159
+ "name" : "sin",
160
+ "type" : "MultiArray"
161
+ },
162
+ {
163
+ "hasShapeFlexibility" : "0",
164
+ "isOptional" : "0",
165
+ "dataType" : "Float16",
166
+ "formattedType" : "MultiArray (Float16 1 × 1 × 32 × 1024)",
167
+ "shortDescription" : "",
168
+ "shape" : "[1, 1, 32, 1024]",
169
+ "name" : "attention_mask",
170
+ "type" : "MultiArray"
171
+ },
172
+ {
173
+ "hasShapeFlexibility" : "0",
174
+ "isOptional" : "0",
175
+ "dataType" : "Int32",
176
+ "formattedType" : "MultiArray (Int32 1)",
177
+ "shortDescription" : "",
178
+ "shape" : "[1]",
179
+ "name" : "n_past",
180
+ "type" : "MultiArray"
181
+ }
182
+ ],
183
+ "computePrecision" : "Mixed (Float16, Int32)",
184
+ "storagePrecision" : "Float16",
185
+ "stateSchema" : [
186
+ {
187
+ "dataType" : "Float16",
188
+ "isOptional" : "0",
189
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
190
+ "shortDescription" : "",
191
+ "shape" : "[4, 8, 1024, 128]",
192
+ "name" : "key_cache",
193
+ "type" : "State"
194
+ },
195
+ {
196
+ "dataType" : "Float16",
197
+ "isOptional" : "0",
198
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
199
+ "shortDescription" : "",
200
+ "shape" : "[4, 8, 1024, 128]",
201
+ "name" : "value_cache",
202
+ "type" : "State"
203
+ }
204
+ ],
205
+ "outputSchema" : [
206
+ {
207
+ "hasShapeFlexibility" : "0",
208
+ "isOptional" : "0",
209
+ "dataType" : "Float16",
210
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
211
+ "shortDescription" : "",
212
+ "shape" : "[1, 32, 2560]",
213
+ "name" : "output",
214
+ "type" : "MultiArray"
215
+ }
216
+ ],
217
+ "name" : "ar32",
218
+ "mlProgramOperationTypeHistogram" : {
219
+ "Ios18.readState" : 10,
220
+ "Ios18.conv" : 28,
221
+ "Ios18.expandDims" : 26,
222
+ "Ios18.matmul" : 8,
223
+ "Ios18.concat" : 29,
224
+ "Ios18.add" : 21,
225
+ "Ios18.silu" : 4,
226
+ "Ios18.sliceUpdate" : 8,
227
+ "Ios18.sliceByIndex" : 40,
228
+ "Ios18.layerNorm" : 16,
229
+ "Ios18.transpose" : 36,
230
+ "Tile" : 8,
231
+ "Ios18.writeState" : 8,
232
+ "Ios18.reshape" : 32,
233
+ "Ios18.softmax" : 4,
234
+ "Ios18.squeeze" : 4,
235
+ "Ios18.mul" : 64
236
+ }
237
+ }
238
+ ],
239
+ "mlProgramOperationTypeHistogram" : {
240
+ "Ios18.readState" : 10,
241
+ "Ios18.conv" : 28,
242
+ "Ios18.expandDims" : 26,
243
+ "Ios18.matmul" : 8,
244
+ "Ios18.concat" : 29,
245
+ "Ios18.add" : 21,
246
+ "Ios18.silu" : 4,
247
+ "Ios18.sliceUpdate" : 8,
248
+ "Ios18.sliceByIndex" : 40,
249
+ "Ios18.layerNorm" : 16,
250
+ "Ios18.transpose" : 36,
251
+ "Tile" : 8,
252
+ "Ios18.writeState" : 8,
253
+ "Ios18.reshape" : 32,
254
+ "Ios18.softmax" : 4,
255
+ "Ios18.squeeze" : 4,
256
+ "Ios18.mul" : 64
257
+ },
258
+ "isUpdatable" : "0",
259
+ "stateSchema" : [
260
+ {
261
+ "dataType" : "Float16",
262
+ "isOptional" : "0",
263
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
264
+ "shortDescription" : "",
265
+ "shape" : "[4, 8, 1024, 128]",
266
+ "name" : "key_cache",
267
+ "type" : "State"
268
+ },
269
+ {
270
+ "dataType" : "Float16",
271
+ "isOptional" : "0",
272
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
273
+ "shortDescription" : "",
274
+ "shape" : "[4, 8, 1024, 128]",
275
+ "name" : "value_cache",
276
+ "type" : "State"
277
+ }
278
+ ],
279
+ "availability" : {
280
+ "macOS" : "15.0",
281
+ "tvOS" : "18.0",
282
+ "visionOS" : "2.0",
283
+ "watchOS" : "11.0",
284
+ "iOS" : "18.0",
285
+ "macCatalyst" : "18.0"
286
+ },
287
+ "computePrecision" : "Mixed (Float16, Int32)",
288
+ "modelType" : {
289
+ "name" : "MLModelType_mlProgram"
290
+ },
291
+ "inputSchema" : [
292
+ {
293
+ "hasShapeFlexibility" : "0",
294
+ "isOptional" : "0",
295
+ "dataType" : "Float16",
296
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
297
+ "shortDescription" : "",
298
+ "shape" : "[1, 1, 2560]",
299
+ "name" : "input_embedding",
300
+ "type" : "MultiArray"
301
+ },
302
+ {
303
+ "hasShapeFlexibility" : "0",
304
+ "isOptional" : "0",
305
+ "dataType" : "Float16",
306
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
307
+ "shortDescription" : "",
308
+ "shape" : "[1, 1, 128]",
309
+ "name" : "cos",
310
+ "type" : "MultiArray"
311
+ },
312
+ {
313
+ "hasShapeFlexibility" : "0",
314
+ "isOptional" : "0",
315
+ "dataType" : "Float16",
316
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
317
+ "shortDescription" : "",
318
+ "shape" : "[1, 1, 128]",
319
+ "name" : "sin",
320
+ "type" : "MultiArray"
321
+ },
322
+ {
323
+ "hasShapeFlexibility" : "0",
324
+ "isOptional" : "0",
325
+ "dataType" : "Float16",
326
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
327
+ "shortDescription" : "",
328
+ "shape" : "[1, 1, 1, 1024]",
329
+ "name" : "attention_mask",
330
+ "type" : "MultiArray"
331
+ },
332
+ {
333
+ "hasShapeFlexibility" : "0",
334
+ "isOptional" : "0",
335
+ "dataType" : "Int32",
336
+ "formattedType" : "MultiArray (Int32 1)",
337
+ "shortDescription" : "",
338
+ "shape" : "[1]",
339
+ "name" : "n_past",
340
+ "type" : "MultiArray"
341
+ }
342
+ ],
343
+ "defaultFunctionName" : "ar1",
344
+ "generatedClassName" : "qwen3_4B_part4_ar1_32",
345
+ "userDefinedMetadata" : {
346
+
347
+ },
348
+ "method" : "predict"
349
+ }
350
+ ]
qwen3-4B-part4-ar1-32.mlmodelc/model.mil ADDED
The diff for this file is too large to render. See raw diff
 
qwen3-4B-part4-ar1-32.mlmodelc/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7977571ed7d69ab512e583cb648aa097bfde2df4670982d91b7f9b0c868bb5e
3
+ size 807449408
qwen3-4B-part5-ar1-32.mlmodelc/analytics/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:846768a9d33810840860dbf42fd49f635d2f5ad3fdb34610e80e276241f7dd90
3
+ size 243
qwen3-4B-part5-ar1-32.mlmodelc/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6181f4eb9eec7d084dc9b7c68b9747a75634e25e74b39b9c3aa26dc526d2c5c
3
+ size 577
qwen3-4B-part5-ar1-32.mlmodelc/metadata.json ADDED
@@ -0,0 +1,350 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "metadataOutputVersion" : "3.0",
4
+ "storagePrecision" : "Float16",
5
+ "outputSchema" : [
6
+ {
7
+ "hasShapeFlexibility" : "0",
8
+ "isOptional" : "0",
9
+ "dataType" : "Float16",
10
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
11
+ "shortDescription" : "",
12
+ "shape" : "[1, 1, 2560]",
13
+ "name" : "output",
14
+ "type" : "MultiArray"
15
+ }
16
+ ],
17
+ "modelParameters" : [
18
+
19
+ ],
20
+ "specificationVersion" : 9,
21
+ "functions" : [
22
+ {
23
+ "inputSchema" : [
24
+ {
25
+ "hasShapeFlexibility" : "0",
26
+ "isOptional" : "0",
27
+ "dataType" : "Float16",
28
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
29
+ "shortDescription" : "",
30
+ "shape" : "[1, 1, 2560]",
31
+ "name" : "input_embedding",
32
+ "type" : "MultiArray"
33
+ },
34
+ {
35
+ "hasShapeFlexibility" : "0",
36
+ "isOptional" : "0",
37
+ "dataType" : "Float16",
38
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
39
+ "shortDescription" : "",
40
+ "shape" : "[1, 1, 128]",
41
+ "name" : "cos",
42
+ "type" : "MultiArray"
43
+ },
44
+ {
45
+ "hasShapeFlexibility" : "0",
46
+ "isOptional" : "0",
47
+ "dataType" : "Float16",
48
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
49
+ "shortDescription" : "",
50
+ "shape" : "[1, 1, 128]",
51
+ "name" : "sin",
52
+ "type" : "MultiArray"
53
+ },
54
+ {
55
+ "hasShapeFlexibility" : "0",
56
+ "isOptional" : "0",
57
+ "dataType" : "Float16",
58
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
59
+ "shortDescription" : "",
60
+ "shape" : "[1, 1, 1, 1024]",
61
+ "name" : "attention_mask",
62
+ "type" : "MultiArray"
63
+ },
64
+ {
65
+ "hasShapeFlexibility" : "0",
66
+ "isOptional" : "0",
67
+ "dataType" : "Int32",
68
+ "formattedType" : "MultiArray (Int32 1)",
69
+ "shortDescription" : "",
70
+ "shape" : "[1]",
71
+ "name" : "n_past",
72
+ "type" : "MultiArray"
73
+ }
74
+ ],
75
+ "computePrecision" : "Mixed (Float16, Int32)",
76
+ "storagePrecision" : "Float16",
77
+ "stateSchema" : [
78
+ {
79
+ "dataType" : "Float16",
80
+ "isOptional" : "0",
81
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
82
+ "shortDescription" : "",
83
+ "shape" : "[4, 8, 1024, 128]",
84
+ "name" : "key_cache",
85
+ "type" : "State"
86
+ },
87
+ {
88
+ "dataType" : "Float16",
89
+ "isOptional" : "0",
90
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
91
+ "shortDescription" : "",
92
+ "shape" : "[4, 8, 1024, 128]",
93
+ "name" : "value_cache",
94
+ "type" : "State"
95
+ }
96
+ ],
97
+ "outputSchema" : [
98
+ {
99
+ "hasShapeFlexibility" : "0",
100
+ "isOptional" : "0",
101
+ "dataType" : "Float16",
102
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
103
+ "shortDescription" : "",
104
+ "shape" : "[1, 1, 2560]",
105
+ "name" : "output",
106
+ "type" : "MultiArray"
107
+ }
108
+ ],
109
+ "name" : "ar1",
110
+ "mlProgramOperationTypeHistogram" : {
111
+ "Ios18.readState" : 10,
112
+ "Ios18.conv" : 28,
113
+ "Ios18.expandDims" : 26,
114
+ "Ios18.matmul" : 8,
115
+ "Ios18.concat" : 29,
116
+ "Ios18.add" : 21,
117
+ "Ios18.silu" : 4,
118
+ "Ios18.sliceUpdate" : 8,
119
+ "Ios18.sliceByIndex" : 40,
120
+ "Ios18.layerNorm" : 16,
121
+ "Ios18.transpose" : 36,
122
+ "Tile" : 8,
123
+ "Ios18.writeState" : 8,
124
+ "Ios18.reshape" : 32,
125
+ "Ios18.softmax" : 4,
126
+ "Ios18.squeeze" : 4,
127
+ "Ios18.mul" : 64
128
+ }
129
+ },
130
+ {
131
+ "inputSchema" : [
132
+ {
133
+ "hasShapeFlexibility" : "0",
134
+ "isOptional" : "0",
135
+ "dataType" : "Float16",
136
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
137
+ "shortDescription" : "",
138
+ "shape" : "[1, 32, 2560]",
139
+ "name" : "input_embedding",
140
+ "type" : "MultiArray"
141
+ },
142
+ {
143
+ "hasShapeFlexibility" : "0",
144
+ "isOptional" : "0",
145
+ "dataType" : "Float16",
146
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
147
+ "shortDescription" : "",
148
+ "shape" : "[1, 32, 128]",
149
+ "name" : "cos",
150
+ "type" : "MultiArray"
151
+ },
152
+ {
153
+ "hasShapeFlexibility" : "0",
154
+ "isOptional" : "0",
155
+ "dataType" : "Float16",
156
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
157
+ "shortDescription" : "",
158
+ "shape" : "[1, 32, 128]",
159
+ "name" : "sin",
160
+ "type" : "MultiArray"
161
+ },
162
+ {
163
+ "hasShapeFlexibility" : "0",
164
+ "isOptional" : "0",
165
+ "dataType" : "Float16",
166
+ "formattedType" : "MultiArray (Float16 1 × 1 × 32 × 1024)",
167
+ "shortDescription" : "",
168
+ "shape" : "[1, 1, 32, 1024]",
169
+ "name" : "attention_mask",
170
+ "type" : "MultiArray"
171
+ },
172
+ {
173
+ "hasShapeFlexibility" : "0",
174
+ "isOptional" : "0",
175
+ "dataType" : "Int32",
176
+ "formattedType" : "MultiArray (Int32 1)",
177
+ "shortDescription" : "",
178
+ "shape" : "[1]",
179
+ "name" : "n_past",
180
+ "type" : "MultiArray"
181
+ }
182
+ ],
183
+ "computePrecision" : "Mixed (Float16, Int32)",
184
+ "storagePrecision" : "Float16",
185
+ "stateSchema" : [
186
+ {
187
+ "dataType" : "Float16",
188
+ "isOptional" : "0",
189
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
190
+ "shortDescription" : "",
191
+ "shape" : "[4, 8, 1024, 128]",
192
+ "name" : "key_cache",
193
+ "type" : "State"
194
+ },
195
+ {
196
+ "dataType" : "Float16",
197
+ "isOptional" : "0",
198
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
199
+ "shortDescription" : "",
200
+ "shape" : "[4, 8, 1024, 128]",
201
+ "name" : "value_cache",
202
+ "type" : "State"
203
+ }
204
+ ],
205
+ "outputSchema" : [
206
+ {
207
+ "hasShapeFlexibility" : "0",
208
+ "isOptional" : "0",
209
+ "dataType" : "Float16",
210
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
211
+ "shortDescription" : "",
212
+ "shape" : "[1, 32, 2560]",
213
+ "name" : "output",
214
+ "type" : "MultiArray"
215
+ }
216
+ ],
217
+ "name" : "ar32",
218
+ "mlProgramOperationTypeHistogram" : {
219
+ "Ios18.readState" : 10,
220
+ "Ios18.conv" : 28,
221
+ "Ios18.expandDims" : 26,
222
+ "Ios18.matmul" : 8,
223
+ "Ios18.concat" : 29,
224
+ "Ios18.add" : 21,
225
+ "Ios18.silu" : 4,
226
+ "Ios18.sliceUpdate" : 8,
227
+ "Ios18.sliceByIndex" : 40,
228
+ "Ios18.layerNorm" : 16,
229
+ "Ios18.transpose" : 36,
230
+ "Tile" : 8,
231
+ "Ios18.writeState" : 8,
232
+ "Ios18.reshape" : 32,
233
+ "Ios18.softmax" : 4,
234
+ "Ios18.squeeze" : 4,
235
+ "Ios18.mul" : 64
236
+ }
237
+ }
238
+ ],
239
+ "mlProgramOperationTypeHistogram" : {
240
+ "Ios18.readState" : 10,
241
+ "Ios18.conv" : 28,
242
+ "Ios18.expandDims" : 26,
243
+ "Ios18.matmul" : 8,
244
+ "Ios18.concat" : 29,
245
+ "Ios18.add" : 21,
246
+ "Ios18.silu" : 4,
247
+ "Ios18.sliceUpdate" : 8,
248
+ "Ios18.sliceByIndex" : 40,
249
+ "Ios18.layerNorm" : 16,
250
+ "Ios18.transpose" : 36,
251
+ "Tile" : 8,
252
+ "Ios18.writeState" : 8,
253
+ "Ios18.reshape" : 32,
254
+ "Ios18.softmax" : 4,
255
+ "Ios18.squeeze" : 4,
256
+ "Ios18.mul" : 64
257
+ },
258
+ "isUpdatable" : "0",
259
+ "stateSchema" : [
260
+ {
261
+ "dataType" : "Float16",
262
+ "isOptional" : "0",
263
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
264
+ "shortDescription" : "",
265
+ "shape" : "[4, 8, 1024, 128]",
266
+ "name" : "key_cache",
267
+ "type" : "State"
268
+ },
269
+ {
270
+ "dataType" : "Float16",
271
+ "isOptional" : "0",
272
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
273
+ "shortDescription" : "",
274
+ "shape" : "[4, 8, 1024, 128]",
275
+ "name" : "value_cache",
276
+ "type" : "State"
277
+ }
278
+ ],
279
+ "availability" : {
280
+ "macOS" : "15.0",
281
+ "tvOS" : "18.0",
282
+ "visionOS" : "2.0",
283
+ "watchOS" : "11.0",
284
+ "iOS" : "18.0",
285
+ "macCatalyst" : "18.0"
286
+ },
287
+ "computePrecision" : "Mixed (Float16, Int32)",
288
+ "modelType" : {
289
+ "name" : "MLModelType_mlProgram"
290
+ },
291
+ "inputSchema" : [
292
+ {
293
+ "hasShapeFlexibility" : "0",
294
+ "isOptional" : "0",
295
+ "dataType" : "Float16",
296
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
297
+ "shortDescription" : "",
298
+ "shape" : "[1, 1, 2560]",
299
+ "name" : "input_embedding",
300
+ "type" : "MultiArray"
301
+ },
302
+ {
303
+ "hasShapeFlexibility" : "0",
304
+ "isOptional" : "0",
305
+ "dataType" : "Float16",
306
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
307
+ "shortDescription" : "",
308
+ "shape" : "[1, 1, 128]",
309
+ "name" : "cos",
310
+ "type" : "MultiArray"
311
+ },
312
+ {
313
+ "hasShapeFlexibility" : "0",
314
+ "isOptional" : "0",
315
+ "dataType" : "Float16",
316
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
317
+ "shortDescription" : "",
318
+ "shape" : "[1, 1, 128]",
319
+ "name" : "sin",
320
+ "type" : "MultiArray"
321
+ },
322
+ {
323
+ "hasShapeFlexibility" : "0",
324
+ "isOptional" : "0",
325
+ "dataType" : "Float16",
326
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
327
+ "shortDescription" : "",
328
+ "shape" : "[1, 1, 1, 1024]",
329
+ "name" : "attention_mask",
330
+ "type" : "MultiArray"
331
+ },
332
+ {
333
+ "hasShapeFlexibility" : "0",
334
+ "isOptional" : "0",
335
+ "dataType" : "Int32",
336
+ "formattedType" : "MultiArray (Int32 1)",
337
+ "shortDescription" : "",
338
+ "shape" : "[1]",
339
+ "name" : "n_past",
340
+ "type" : "MultiArray"
341
+ }
342
+ ],
343
+ "defaultFunctionName" : "ar1",
344
+ "generatedClassName" : "qwen3_4B_part5_ar1_32",
345
+ "userDefinedMetadata" : {
346
+
347
+ },
348
+ "method" : "predict"
349
+ }
350
+ ]
qwen3-4B-part5-ar1-32.mlmodelc/model.mil ADDED
The diff for this file is too large to render. See raw diff
 
qwen3-4B-part5-ar1-32.mlmodelc/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:765c5e77a948fad39e11b463d0adc350111145fe8f5fec9f591b13fe403871fa
3
+ size 807449408
qwen3-4B-part6-ar1-32.mlmodelc/analytics/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:846768a9d33810840860dbf42fd49f635d2f5ad3fdb34610e80e276241f7dd90
3
+ size 243
qwen3-4B-part6-ar1-32.mlmodelc/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6181f4eb9eec7d084dc9b7c68b9747a75634e25e74b39b9c3aa26dc526d2c5c
3
+ size 577
qwen3-4B-part6-ar1-32.mlmodelc/metadata.json ADDED
@@ -0,0 +1,350 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "metadataOutputVersion" : "3.0",
4
+ "storagePrecision" : "Float16",
5
+ "outputSchema" : [
6
+ {
7
+ "hasShapeFlexibility" : "0",
8
+ "isOptional" : "0",
9
+ "dataType" : "Float16",
10
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
11
+ "shortDescription" : "",
12
+ "shape" : "[1, 1, 2560]",
13
+ "name" : "output",
14
+ "type" : "MultiArray"
15
+ }
16
+ ],
17
+ "modelParameters" : [
18
+
19
+ ],
20
+ "specificationVersion" : 9,
21
+ "functions" : [
22
+ {
23
+ "inputSchema" : [
24
+ {
25
+ "hasShapeFlexibility" : "0",
26
+ "isOptional" : "0",
27
+ "dataType" : "Float16",
28
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
29
+ "shortDescription" : "",
30
+ "shape" : "[1, 1, 2560]",
31
+ "name" : "input_embedding",
32
+ "type" : "MultiArray"
33
+ },
34
+ {
35
+ "hasShapeFlexibility" : "0",
36
+ "isOptional" : "0",
37
+ "dataType" : "Float16",
38
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
39
+ "shortDescription" : "",
40
+ "shape" : "[1, 1, 128]",
41
+ "name" : "cos",
42
+ "type" : "MultiArray"
43
+ },
44
+ {
45
+ "hasShapeFlexibility" : "0",
46
+ "isOptional" : "0",
47
+ "dataType" : "Float16",
48
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
49
+ "shortDescription" : "",
50
+ "shape" : "[1, 1, 128]",
51
+ "name" : "sin",
52
+ "type" : "MultiArray"
53
+ },
54
+ {
55
+ "hasShapeFlexibility" : "0",
56
+ "isOptional" : "0",
57
+ "dataType" : "Float16",
58
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
59
+ "shortDescription" : "",
60
+ "shape" : "[1, 1, 1, 1024]",
61
+ "name" : "attention_mask",
62
+ "type" : "MultiArray"
63
+ },
64
+ {
65
+ "hasShapeFlexibility" : "0",
66
+ "isOptional" : "0",
67
+ "dataType" : "Int32",
68
+ "formattedType" : "MultiArray (Int32 1)",
69
+ "shortDescription" : "",
70
+ "shape" : "[1]",
71
+ "name" : "n_past",
72
+ "type" : "MultiArray"
73
+ }
74
+ ],
75
+ "computePrecision" : "Mixed (Float16, Int32)",
76
+ "storagePrecision" : "Float16",
77
+ "stateSchema" : [
78
+ {
79
+ "dataType" : "Float16",
80
+ "isOptional" : "0",
81
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
82
+ "shortDescription" : "",
83
+ "shape" : "[4, 8, 1024, 128]",
84
+ "name" : "key_cache",
85
+ "type" : "State"
86
+ },
87
+ {
88
+ "dataType" : "Float16",
89
+ "isOptional" : "0",
90
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
91
+ "shortDescription" : "",
92
+ "shape" : "[4, 8, 1024, 128]",
93
+ "name" : "value_cache",
94
+ "type" : "State"
95
+ }
96
+ ],
97
+ "outputSchema" : [
98
+ {
99
+ "hasShapeFlexibility" : "0",
100
+ "isOptional" : "0",
101
+ "dataType" : "Float16",
102
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
103
+ "shortDescription" : "",
104
+ "shape" : "[1, 1, 2560]",
105
+ "name" : "output",
106
+ "type" : "MultiArray"
107
+ }
108
+ ],
109
+ "name" : "ar1",
110
+ "mlProgramOperationTypeHistogram" : {
111
+ "Ios18.readState" : 10,
112
+ "Ios18.conv" : 28,
113
+ "Ios18.expandDims" : 26,
114
+ "Ios18.matmul" : 8,
115
+ "Ios18.concat" : 29,
116
+ "Ios18.add" : 21,
117
+ "Ios18.silu" : 4,
118
+ "Ios18.sliceUpdate" : 8,
119
+ "Ios18.sliceByIndex" : 40,
120
+ "Ios18.layerNorm" : 16,
121
+ "Ios18.transpose" : 36,
122
+ "Tile" : 8,
123
+ "Ios18.writeState" : 8,
124
+ "Ios18.reshape" : 32,
125
+ "Ios18.softmax" : 4,
126
+ "Ios18.squeeze" : 4,
127
+ "Ios18.mul" : 64
128
+ }
129
+ },
130
+ {
131
+ "inputSchema" : [
132
+ {
133
+ "hasShapeFlexibility" : "0",
134
+ "isOptional" : "0",
135
+ "dataType" : "Float16",
136
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
137
+ "shortDescription" : "",
138
+ "shape" : "[1, 32, 2560]",
139
+ "name" : "input_embedding",
140
+ "type" : "MultiArray"
141
+ },
142
+ {
143
+ "hasShapeFlexibility" : "0",
144
+ "isOptional" : "0",
145
+ "dataType" : "Float16",
146
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
147
+ "shortDescription" : "",
148
+ "shape" : "[1, 32, 128]",
149
+ "name" : "cos",
150
+ "type" : "MultiArray"
151
+ },
152
+ {
153
+ "hasShapeFlexibility" : "0",
154
+ "isOptional" : "0",
155
+ "dataType" : "Float16",
156
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
157
+ "shortDescription" : "",
158
+ "shape" : "[1, 32, 128]",
159
+ "name" : "sin",
160
+ "type" : "MultiArray"
161
+ },
162
+ {
163
+ "hasShapeFlexibility" : "0",
164
+ "isOptional" : "0",
165
+ "dataType" : "Float16",
166
+ "formattedType" : "MultiArray (Float16 1 × 1 × 32 × 1024)",
167
+ "shortDescription" : "",
168
+ "shape" : "[1, 1, 32, 1024]",
169
+ "name" : "attention_mask",
170
+ "type" : "MultiArray"
171
+ },
172
+ {
173
+ "hasShapeFlexibility" : "0",
174
+ "isOptional" : "0",
175
+ "dataType" : "Int32",
176
+ "formattedType" : "MultiArray (Int32 1)",
177
+ "shortDescription" : "",
178
+ "shape" : "[1]",
179
+ "name" : "n_past",
180
+ "type" : "MultiArray"
181
+ }
182
+ ],
183
+ "computePrecision" : "Mixed (Float16, Int32)",
184
+ "storagePrecision" : "Float16",
185
+ "stateSchema" : [
186
+ {
187
+ "dataType" : "Float16",
188
+ "isOptional" : "0",
189
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
190
+ "shortDescription" : "",
191
+ "shape" : "[4, 8, 1024, 128]",
192
+ "name" : "key_cache",
193
+ "type" : "State"
194
+ },
195
+ {
196
+ "dataType" : "Float16",
197
+ "isOptional" : "0",
198
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
199
+ "shortDescription" : "",
200
+ "shape" : "[4, 8, 1024, 128]",
201
+ "name" : "value_cache",
202
+ "type" : "State"
203
+ }
204
+ ],
205
+ "outputSchema" : [
206
+ {
207
+ "hasShapeFlexibility" : "0",
208
+ "isOptional" : "0",
209
+ "dataType" : "Float16",
210
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
211
+ "shortDescription" : "",
212
+ "shape" : "[1, 32, 2560]",
213
+ "name" : "output",
214
+ "type" : "MultiArray"
215
+ }
216
+ ],
217
+ "name" : "ar32",
218
+ "mlProgramOperationTypeHistogram" : {
219
+ "Ios18.readState" : 10,
220
+ "Ios18.conv" : 28,
221
+ "Ios18.expandDims" : 26,
222
+ "Ios18.matmul" : 8,
223
+ "Ios18.concat" : 29,
224
+ "Ios18.add" : 21,
225
+ "Ios18.silu" : 4,
226
+ "Ios18.sliceUpdate" : 8,
227
+ "Ios18.sliceByIndex" : 40,
228
+ "Ios18.layerNorm" : 16,
229
+ "Ios18.transpose" : 36,
230
+ "Tile" : 8,
231
+ "Ios18.writeState" : 8,
232
+ "Ios18.reshape" : 32,
233
+ "Ios18.softmax" : 4,
234
+ "Ios18.squeeze" : 4,
235
+ "Ios18.mul" : 64
236
+ }
237
+ }
238
+ ],
239
+ "mlProgramOperationTypeHistogram" : {
240
+ "Ios18.readState" : 10,
241
+ "Ios18.conv" : 28,
242
+ "Ios18.expandDims" : 26,
243
+ "Ios18.matmul" : 8,
244
+ "Ios18.concat" : 29,
245
+ "Ios18.add" : 21,
246
+ "Ios18.silu" : 4,
247
+ "Ios18.sliceUpdate" : 8,
248
+ "Ios18.sliceByIndex" : 40,
249
+ "Ios18.layerNorm" : 16,
250
+ "Ios18.transpose" : 36,
251
+ "Tile" : 8,
252
+ "Ios18.writeState" : 8,
253
+ "Ios18.reshape" : 32,
254
+ "Ios18.softmax" : 4,
255
+ "Ios18.squeeze" : 4,
256
+ "Ios18.mul" : 64
257
+ },
258
+ "isUpdatable" : "0",
259
+ "stateSchema" : [
260
+ {
261
+ "dataType" : "Float16",
262
+ "isOptional" : "0",
263
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
264
+ "shortDescription" : "",
265
+ "shape" : "[4, 8, 1024, 128]",
266
+ "name" : "key_cache",
267
+ "type" : "State"
268
+ },
269
+ {
270
+ "dataType" : "Float16",
271
+ "isOptional" : "0",
272
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
273
+ "shortDescription" : "",
274
+ "shape" : "[4, 8, 1024, 128]",
275
+ "name" : "value_cache",
276
+ "type" : "State"
277
+ }
278
+ ],
279
+ "availability" : {
280
+ "macOS" : "15.0",
281
+ "tvOS" : "18.0",
282
+ "visionOS" : "2.0",
283
+ "watchOS" : "11.0",
284
+ "iOS" : "18.0",
285
+ "macCatalyst" : "18.0"
286
+ },
287
+ "computePrecision" : "Mixed (Float16, Int32)",
288
+ "modelType" : {
289
+ "name" : "MLModelType_mlProgram"
290
+ },
291
+ "inputSchema" : [
292
+ {
293
+ "hasShapeFlexibility" : "0",
294
+ "isOptional" : "0",
295
+ "dataType" : "Float16",
296
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
297
+ "shortDescription" : "",
298
+ "shape" : "[1, 1, 2560]",
299
+ "name" : "input_embedding",
300
+ "type" : "MultiArray"
301
+ },
302
+ {
303
+ "hasShapeFlexibility" : "0",
304
+ "isOptional" : "0",
305
+ "dataType" : "Float16",
306
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
307
+ "shortDescription" : "",
308
+ "shape" : "[1, 1, 128]",
309
+ "name" : "cos",
310
+ "type" : "MultiArray"
311
+ },
312
+ {
313
+ "hasShapeFlexibility" : "0",
314
+ "isOptional" : "0",
315
+ "dataType" : "Float16",
316
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
317
+ "shortDescription" : "",
318
+ "shape" : "[1, 1, 128]",
319
+ "name" : "sin",
320
+ "type" : "MultiArray"
321
+ },
322
+ {
323
+ "hasShapeFlexibility" : "0",
324
+ "isOptional" : "0",
325
+ "dataType" : "Float16",
326
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
327
+ "shortDescription" : "",
328
+ "shape" : "[1, 1, 1, 1024]",
329
+ "name" : "attention_mask",
330
+ "type" : "MultiArray"
331
+ },
332
+ {
333
+ "hasShapeFlexibility" : "0",
334
+ "isOptional" : "0",
335
+ "dataType" : "Int32",
336
+ "formattedType" : "MultiArray (Int32 1)",
337
+ "shortDescription" : "",
338
+ "shape" : "[1]",
339
+ "name" : "n_past",
340
+ "type" : "MultiArray"
341
+ }
342
+ ],
343
+ "defaultFunctionName" : "ar1",
344
+ "generatedClassName" : "qwen3_4B_part6_ar1_32",
345
+ "userDefinedMetadata" : {
346
+
347
+ },
348
+ "method" : "predict"
349
+ }
350
+ ]
qwen3-4B-part6-ar1-32.mlmodelc/model.mil ADDED
The diff for this file is too large to render. See raw diff
 
qwen3-4B-part6-ar1-32.mlmodelc/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:36b8e238e31f5dba01845f2132bb54a3ccdf5c92b18adcd687a3aeb90d63798a
3
+ size 807449408
qwen3-4B-part7-ar1-32.mlmodelc/analytics/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:846768a9d33810840860dbf42fd49f635d2f5ad3fdb34610e80e276241f7dd90
3
+ size 243
qwen3-4B-part7-ar1-32.mlmodelc/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6181f4eb9eec7d084dc9b7c68b9747a75634e25e74b39b9c3aa26dc526d2c5c
3
+ size 577
qwen3-4B-part7-ar1-32.mlmodelc/metadata.json ADDED
@@ -0,0 +1,350 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "metadataOutputVersion" : "3.0",
4
+ "storagePrecision" : "Float16",
5
+ "outputSchema" : [
6
+ {
7
+ "hasShapeFlexibility" : "0",
8
+ "isOptional" : "0",
9
+ "dataType" : "Float16",
10
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
11
+ "shortDescription" : "",
12
+ "shape" : "[1, 1, 2560]",
13
+ "name" : "output",
14
+ "type" : "MultiArray"
15
+ }
16
+ ],
17
+ "modelParameters" : [
18
+
19
+ ],
20
+ "specificationVersion" : 9,
21
+ "functions" : [
22
+ {
23
+ "inputSchema" : [
24
+ {
25
+ "hasShapeFlexibility" : "0",
26
+ "isOptional" : "0",
27
+ "dataType" : "Float16",
28
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
29
+ "shortDescription" : "",
30
+ "shape" : "[1, 1, 2560]",
31
+ "name" : "input_embedding",
32
+ "type" : "MultiArray"
33
+ },
34
+ {
35
+ "hasShapeFlexibility" : "0",
36
+ "isOptional" : "0",
37
+ "dataType" : "Float16",
38
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
39
+ "shortDescription" : "",
40
+ "shape" : "[1, 1, 128]",
41
+ "name" : "cos",
42
+ "type" : "MultiArray"
43
+ },
44
+ {
45
+ "hasShapeFlexibility" : "0",
46
+ "isOptional" : "0",
47
+ "dataType" : "Float16",
48
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
49
+ "shortDescription" : "",
50
+ "shape" : "[1, 1, 128]",
51
+ "name" : "sin",
52
+ "type" : "MultiArray"
53
+ },
54
+ {
55
+ "hasShapeFlexibility" : "0",
56
+ "isOptional" : "0",
57
+ "dataType" : "Float16",
58
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
59
+ "shortDescription" : "",
60
+ "shape" : "[1, 1, 1, 1024]",
61
+ "name" : "attention_mask",
62
+ "type" : "MultiArray"
63
+ },
64
+ {
65
+ "hasShapeFlexibility" : "0",
66
+ "isOptional" : "0",
67
+ "dataType" : "Int32",
68
+ "formattedType" : "MultiArray (Int32 1)",
69
+ "shortDescription" : "",
70
+ "shape" : "[1]",
71
+ "name" : "n_past",
72
+ "type" : "MultiArray"
73
+ }
74
+ ],
75
+ "computePrecision" : "Mixed (Float16, Int32)",
76
+ "storagePrecision" : "Float16",
77
+ "stateSchema" : [
78
+ {
79
+ "dataType" : "Float16",
80
+ "isOptional" : "0",
81
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
82
+ "shortDescription" : "",
83
+ "shape" : "[4, 8, 1024, 128]",
84
+ "name" : "key_cache",
85
+ "type" : "State"
86
+ },
87
+ {
88
+ "dataType" : "Float16",
89
+ "isOptional" : "0",
90
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
91
+ "shortDescription" : "",
92
+ "shape" : "[4, 8, 1024, 128]",
93
+ "name" : "value_cache",
94
+ "type" : "State"
95
+ }
96
+ ],
97
+ "outputSchema" : [
98
+ {
99
+ "hasShapeFlexibility" : "0",
100
+ "isOptional" : "0",
101
+ "dataType" : "Float16",
102
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
103
+ "shortDescription" : "",
104
+ "shape" : "[1, 1, 2560]",
105
+ "name" : "output",
106
+ "type" : "MultiArray"
107
+ }
108
+ ],
109
+ "name" : "ar1",
110
+ "mlProgramOperationTypeHistogram" : {
111
+ "Ios18.readState" : 10,
112
+ "Ios18.conv" : 28,
113
+ "Ios18.expandDims" : 26,
114
+ "Ios18.matmul" : 8,
115
+ "Ios18.concat" : 29,
116
+ "Ios18.add" : 21,
117
+ "Ios18.silu" : 4,
118
+ "Ios18.sliceUpdate" : 8,
119
+ "Ios18.sliceByIndex" : 40,
120
+ "Ios18.layerNorm" : 16,
121
+ "Ios18.transpose" : 36,
122
+ "Tile" : 8,
123
+ "Ios18.writeState" : 8,
124
+ "Ios18.reshape" : 32,
125
+ "Ios18.softmax" : 4,
126
+ "Ios18.squeeze" : 4,
127
+ "Ios18.mul" : 64
128
+ }
129
+ },
130
+ {
131
+ "inputSchema" : [
132
+ {
133
+ "hasShapeFlexibility" : "0",
134
+ "isOptional" : "0",
135
+ "dataType" : "Float16",
136
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
137
+ "shortDescription" : "",
138
+ "shape" : "[1, 32, 2560]",
139
+ "name" : "input_embedding",
140
+ "type" : "MultiArray"
141
+ },
142
+ {
143
+ "hasShapeFlexibility" : "0",
144
+ "isOptional" : "0",
145
+ "dataType" : "Float16",
146
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
147
+ "shortDescription" : "",
148
+ "shape" : "[1, 32, 128]",
149
+ "name" : "cos",
150
+ "type" : "MultiArray"
151
+ },
152
+ {
153
+ "hasShapeFlexibility" : "0",
154
+ "isOptional" : "0",
155
+ "dataType" : "Float16",
156
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
157
+ "shortDescription" : "",
158
+ "shape" : "[1, 32, 128]",
159
+ "name" : "sin",
160
+ "type" : "MultiArray"
161
+ },
162
+ {
163
+ "hasShapeFlexibility" : "0",
164
+ "isOptional" : "0",
165
+ "dataType" : "Float16",
166
+ "formattedType" : "MultiArray (Float16 1 × 1 × 32 × 1024)",
167
+ "shortDescription" : "",
168
+ "shape" : "[1, 1, 32, 1024]",
169
+ "name" : "attention_mask",
170
+ "type" : "MultiArray"
171
+ },
172
+ {
173
+ "hasShapeFlexibility" : "0",
174
+ "isOptional" : "0",
175
+ "dataType" : "Int32",
176
+ "formattedType" : "MultiArray (Int32 1)",
177
+ "shortDescription" : "",
178
+ "shape" : "[1]",
179
+ "name" : "n_past",
180
+ "type" : "MultiArray"
181
+ }
182
+ ],
183
+ "computePrecision" : "Mixed (Float16, Int32)",
184
+ "storagePrecision" : "Float16",
185
+ "stateSchema" : [
186
+ {
187
+ "dataType" : "Float16",
188
+ "isOptional" : "0",
189
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
190
+ "shortDescription" : "",
191
+ "shape" : "[4, 8, 1024, 128]",
192
+ "name" : "key_cache",
193
+ "type" : "State"
194
+ },
195
+ {
196
+ "dataType" : "Float16",
197
+ "isOptional" : "0",
198
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
199
+ "shortDescription" : "",
200
+ "shape" : "[4, 8, 1024, 128]",
201
+ "name" : "value_cache",
202
+ "type" : "State"
203
+ }
204
+ ],
205
+ "outputSchema" : [
206
+ {
207
+ "hasShapeFlexibility" : "0",
208
+ "isOptional" : "0",
209
+ "dataType" : "Float16",
210
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
211
+ "shortDescription" : "",
212
+ "shape" : "[1, 32, 2560]",
213
+ "name" : "output",
214
+ "type" : "MultiArray"
215
+ }
216
+ ],
217
+ "name" : "ar32",
218
+ "mlProgramOperationTypeHistogram" : {
219
+ "Ios18.readState" : 10,
220
+ "Ios18.conv" : 28,
221
+ "Ios18.expandDims" : 26,
222
+ "Ios18.matmul" : 8,
223
+ "Ios18.concat" : 29,
224
+ "Ios18.add" : 21,
225
+ "Ios18.silu" : 4,
226
+ "Ios18.sliceUpdate" : 8,
227
+ "Ios18.sliceByIndex" : 40,
228
+ "Ios18.layerNorm" : 16,
229
+ "Ios18.transpose" : 36,
230
+ "Tile" : 8,
231
+ "Ios18.writeState" : 8,
232
+ "Ios18.reshape" : 32,
233
+ "Ios18.softmax" : 4,
234
+ "Ios18.squeeze" : 4,
235
+ "Ios18.mul" : 64
236
+ }
237
+ }
238
+ ],
239
+ "mlProgramOperationTypeHistogram" : {
240
+ "Ios18.readState" : 10,
241
+ "Ios18.conv" : 28,
242
+ "Ios18.expandDims" : 26,
243
+ "Ios18.matmul" : 8,
244
+ "Ios18.concat" : 29,
245
+ "Ios18.add" : 21,
246
+ "Ios18.silu" : 4,
247
+ "Ios18.sliceUpdate" : 8,
248
+ "Ios18.sliceByIndex" : 40,
249
+ "Ios18.layerNorm" : 16,
250
+ "Ios18.transpose" : 36,
251
+ "Tile" : 8,
252
+ "Ios18.writeState" : 8,
253
+ "Ios18.reshape" : 32,
254
+ "Ios18.softmax" : 4,
255
+ "Ios18.squeeze" : 4,
256
+ "Ios18.mul" : 64
257
+ },
258
+ "isUpdatable" : "0",
259
+ "stateSchema" : [
260
+ {
261
+ "dataType" : "Float16",
262
+ "isOptional" : "0",
263
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
264
+ "shortDescription" : "",
265
+ "shape" : "[4, 8, 1024, 128]",
266
+ "name" : "key_cache",
267
+ "type" : "State"
268
+ },
269
+ {
270
+ "dataType" : "Float16",
271
+ "isOptional" : "0",
272
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
273
+ "shortDescription" : "",
274
+ "shape" : "[4, 8, 1024, 128]",
275
+ "name" : "value_cache",
276
+ "type" : "State"
277
+ }
278
+ ],
279
+ "availability" : {
280
+ "macOS" : "15.0",
281
+ "tvOS" : "18.0",
282
+ "visionOS" : "2.0",
283
+ "watchOS" : "11.0",
284
+ "iOS" : "18.0",
285
+ "macCatalyst" : "18.0"
286
+ },
287
+ "computePrecision" : "Mixed (Float16, Int32)",
288
+ "modelType" : {
289
+ "name" : "MLModelType_mlProgram"
290
+ },
291
+ "inputSchema" : [
292
+ {
293
+ "hasShapeFlexibility" : "0",
294
+ "isOptional" : "0",
295
+ "dataType" : "Float16",
296
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
297
+ "shortDescription" : "",
298
+ "shape" : "[1, 1, 2560]",
299
+ "name" : "input_embedding",
300
+ "type" : "MultiArray"
301
+ },
302
+ {
303
+ "hasShapeFlexibility" : "0",
304
+ "isOptional" : "0",
305
+ "dataType" : "Float16",
306
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
307
+ "shortDescription" : "",
308
+ "shape" : "[1, 1, 128]",
309
+ "name" : "cos",
310
+ "type" : "MultiArray"
311
+ },
312
+ {
313
+ "hasShapeFlexibility" : "0",
314
+ "isOptional" : "0",
315
+ "dataType" : "Float16",
316
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
317
+ "shortDescription" : "",
318
+ "shape" : "[1, 1, 128]",
319
+ "name" : "sin",
320
+ "type" : "MultiArray"
321
+ },
322
+ {
323
+ "hasShapeFlexibility" : "0",
324
+ "isOptional" : "0",
325
+ "dataType" : "Float16",
326
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
327
+ "shortDescription" : "",
328
+ "shape" : "[1, 1, 1, 1024]",
329
+ "name" : "attention_mask",
330
+ "type" : "MultiArray"
331
+ },
332
+ {
333
+ "hasShapeFlexibility" : "0",
334
+ "isOptional" : "0",
335
+ "dataType" : "Int32",
336
+ "formattedType" : "MultiArray (Int32 1)",
337
+ "shortDescription" : "",
338
+ "shape" : "[1]",
339
+ "name" : "n_past",
340
+ "type" : "MultiArray"
341
+ }
342
+ ],
343
+ "defaultFunctionName" : "ar1",
344
+ "generatedClassName" : "qwen3_4B_part7_ar1_32",
345
+ "userDefinedMetadata" : {
346
+
347
+ },
348
+ "method" : "predict"
349
+ }
350
+ ]
qwen3-4B-part7-ar1-32.mlmodelc/model.mil ADDED
The diff for this file is too large to render. See raw diff
 
qwen3-4B-part7-ar1-32.mlmodelc/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e69387086e7011da544f81fd19fe78362c35cdb1cffafae35869da2fa7679e4
3
+ size 807449408
qwen3-4B-part8-ar1-32.mlmodelc/analytics/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91ebf12407658a60ff5d2617a95eb4eafc4852b20b85ee1baeadf39e332d0c46
3
+ size 243
qwen3-4B-part8-ar1-32.mlmodelc/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c21074d1b46215b9e465ff27a26b4d5493e1f3b88d0cb99f3b49a6ce459f682
3
+ size 579
qwen3-4B-part8-ar1-32.mlmodelc/metadata.json ADDED
@@ -0,0 +1,350 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "metadataOutputVersion" : "3.0",
4
+ "storagePrecision" : "Float16",
5
+ "outputSchema" : [
6
+ {
7
+ "hasShapeFlexibility" : "0",
8
+ "isOptional" : "0",
9
+ "dataType" : "Float16",
10
+ "formattedType" : "MultiArray (Float16 1 × 1 × 151936)",
11
+ "shortDescription" : "",
12
+ "shape" : "[1, 1, 151936]",
13
+ "name" : "output",
14
+ "type" : "MultiArray"
15
+ }
16
+ ],
17
+ "modelParameters" : [
18
+
19
+ ],
20
+ "specificationVersion" : 9,
21
+ "functions" : [
22
+ {
23
+ "inputSchema" : [
24
+ {
25
+ "hasShapeFlexibility" : "0",
26
+ "isOptional" : "0",
27
+ "dataType" : "Float16",
28
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
29
+ "shortDescription" : "",
30
+ "shape" : "[1, 1, 2560]",
31
+ "name" : "input_embedding",
32
+ "type" : "MultiArray"
33
+ },
34
+ {
35
+ "hasShapeFlexibility" : "0",
36
+ "isOptional" : "0",
37
+ "dataType" : "Float16",
38
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
39
+ "shortDescription" : "",
40
+ "shape" : "[1, 1, 128]",
41
+ "name" : "cos",
42
+ "type" : "MultiArray"
43
+ },
44
+ {
45
+ "hasShapeFlexibility" : "0",
46
+ "isOptional" : "0",
47
+ "dataType" : "Float16",
48
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
49
+ "shortDescription" : "",
50
+ "shape" : "[1, 1, 128]",
51
+ "name" : "sin",
52
+ "type" : "MultiArray"
53
+ },
54
+ {
55
+ "hasShapeFlexibility" : "0",
56
+ "isOptional" : "0",
57
+ "dataType" : "Float16",
58
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
59
+ "shortDescription" : "",
60
+ "shape" : "[1, 1, 1, 1024]",
61
+ "name" : "attention_mask",
62
+ "type" : "MultiArray"
63
+ },
64
+ {
65
+ "hasShapeFlexibility" : "0",
66
+ "isOptional" : "0",
67
+ "dataType" : "Int32",
68
+ "formattedType" : "MultiArray (Int32 1)",
69
+ "shortDescription" : "",
70
+ "shape" : "[1]",
71
+ "name" : "n_past",
72
+ "type" : "MultiArray"
73
+ }
74
+ ],
75
+ "computePrecision" : "Mixed (Float16, Int32)",
76
+ "storagePrecision" : "Float16",
77
+ "stateSchema" : [
78
+ {
79
+ "dataType" : "Float16",
80
+ "isOptional" : "0",
81
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
82
+ "shortDescription" : "",
83
+ "shape" : "[4, 8, 1024, 128]",
84
+ "name" : "key_cache",
85
+ "type" : "State"
86
+ },
87
+ {
88
+ "dataType" : "Float16",
89
+ "isOptional" : "0",
90
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
91
+ "shortDescription" : "",
92
+ "shape" : "[4, 8, 1024, 128]",
93
+ "name" : "value_cache",
94
+ "type" : "State"
95
+ }
96
+ ],
97
+ "outputSchema" : [
98
+ {
99
+ "hasShapeFlexibility" : "0",
100
+ "isOptional" : "0",
101
+ "dataType" : "Float16",
102
+ "formattedType" : "MultiArray (Float16 1 × 1 × 151936)",
103
+ "shortDescription" : "",
104
+ "shape" : "[1, 1, 151936]",
105
+ "name" : "output",
106
+ "type" : "MultiArray"
107
+ }
108
+ ],
109
+ "name" : "ar1",
110
+ "mlProgramOperationTypeHistogram" : {
111
+ "Ios18.readState" : 10,
112
+ "Ios18.conv" : 44,
113
+ "Ios18.expandDims" : 27,
114
+ "Ios18.matmul" : 8,
115
+ "Ios18.concat" : 31,
116
+ "Ios18.add" : 21,
117
+ "Ios18.silu" : 4,
118
+ "Ios18.sliceUpdate" : 8,
119
+ "Ios18.sliceByIndex" : 41,
120
+ "Ios18.layerNorm" : 17,
121
+ "Ios18.transpose" : 38,
122
+ "Tile" : 8,
123
+ "Ios18.writeState" : 8,
124
+ "Ios18.reshape" : 32,
125
+ "Ios18.softmax" : 4,
126
+ "Ios18.squeeze" : 20,
127
+ "Ios18.mul" : 66
128
+ }
129
+ },
130
+ {
131
+ "inputSchema" : [
132
+ {
133
+ "hasShapeFlexibility" : "0",
134
+ "isOptional" : "0",
135
+ "dataType" : "Float16",
136
+ "formattedType" : "MultiArray (Float16 1 × 32 × 2560)",
137
+ "shortDescription" : "",
138
+ "shape" : "[1, 32, 2560]",
139
+ "name" : "input_embedding",
140
+ "type" : "MultiArray"
141
+ },
142
+ {
143
+ "hasShapeFlexibility" : "0",
144
+ "isOptional" : "0",
145
+ "dataType" : "Float16",
146
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
147
+ "shortDescription" : "",
148
+ "shape" : "[1, 32, 128]",
149
+ "name" : "cos",
150
+ "type" : "MultiArray"
151
+ },
152
+ {
153
+ "hasShapeFlexibility" : "0",
154
+ "isOptional" : "0",
155
+ "dataType" : "Float16",
156
+ "formattedType" : "MultiArray (Float16 1 × 32 × 128)",
157
+ "shortDescription" : "",
158
+ "shape" : "[1, 32, 128]",
159
+ "name" : "sin",
160
+ "type" : "MultiArray"
161
+ },
162
+ {
163
+ "hasShapeFlexibility" : "0",
164
+ "isOptional" : "0",
165
+ "dataType" : "Float16",
166
+ "formattedType" : "MultiArray (Float16 1 × 1 × 32 × 1024)",
167
+ "shortDescription" : "",
168
+ "shape" : "[1, 1, 32, 1024]",
169
+ "name" : "attention_mask",
170
+ "type" : "MultiArray"
171
+ },
172
+ {
173
+ "hasShapeFlexibility" : "0",
174
+ "isOptional" : "0",
175
+ "dataType" : "Int32",
176
+ "formattedType" : "MultiArray (Int32 1)",
177
+ "shortDescription" : "",
178
+ "shape" : "[1]",
179
+ "name" : "n_past",
180
+ "type" : "MultiArray"
181
+ }
182
+ ],
183
+ "computePrecision" : "Mixed (Float16, Int32)",
184
+ "storagePrecision" : "Float16",
185
+ "stateSchema" : [
186
+ {
187
+ "dataType" : "Float16",
188
+ "isOptional" : "0",
189
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
190
+ "shortDescription" : "",
191
+ "shape" : "[4, 8, 1024, 128]",
192
+ "name" : "key_cache",
193
+ "type" : "State"
194
+ },
195
+ {
196
+ "dataType" : "Float16",
197
+ "isOptional" : "0",
198
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
199
+ "shortDescription" : "",
200
+ "shape" : "[4, 8, 1024, 128]",
201
+ "name" : "value_cache",
202
+ "type" : "State"
203
+ }
204
+ ],
205
+ "outputSchema" : [
206
+ {
207
+ "hasShapeFlexibility" : "0",
208
+ "isOptional" : "0",
209
+ "dataType" : "Float16",
210
+ "formattedType" : "MultiArray (Float16 1 × 32 × 151936)",
211
+ "shortDescription" : "",
212
+ "shape" : "[1, 32, 151936]",
213
+ "name" : "output",
214
+ "type" : "MultiArray"
215
+ }
216
+ ],
217
+ "name" : "ar32",
218
+ "mlProgramOperationTypeHistogram" : {
219
+ "Ios18.readState" : 10,
220
+ "Ios18.conv" : 44,
221
+ "Ios18.expandDims" : 27,
222
+ "Ios18.matmul" : 8,
223
+ "Ios18.concat" : 31,
224
+ "Ios18.add" : 21,
225
+ "Ios18.silu" : 4,
226
+ "Ios18.sliceUpdate" : 8,
227
+ "Ios18.sliceByIndex" : 41,
228
+ "Ios18.layerNorm" : 17,
229
+ "Ios18.transpose" : 38,
230
+ "Tile" : 8,
231
+ "Ios18.writeState" : 8,
232
+ "Ios18.reshape" : 32,
233
+ "Ios18.softmax" : 4,
234
+ "Ios18.squeeze" : 20,
235
+ "Ios18.mul" : 66
236
+ }
237
+ }
238
+ ],
239
+ "mlProgramOperationTypeHistogram" : {
240
+ "Ios18.readState" : 10,
241
+ "Ios18.conv" : 44,
242
+ "Ios18.expandDims" : 27,
243
+ "Ios18.matmul" : 8,
244
+ "Ios18.concat" : 31,
245
+ "Ios18.add" : 21,
246
+ "Ios18.silu" : 4,
247
+ "Ios18.sliceUpdate" : 8,
248
+ "Ios18.sliceByIndex" : 41,
249
+ "Ios18.layerNorm" : 17,
250
+ "Ios18.transpose" : 38,
251
+ "Tile" : 8,
252
+ "Ios18.writeState" : 8,
253
+ "Ios18.reshape" : 32,
254
+ "Ios18.softmax" : 4,
255
+ "Ios18.squeeze" : 20,
256
+ "Ios18.mul" : 66
257
+ },
258
+ "isUpdatable" : "0",
259
+ "stateSchema" : [
260
+ {
261
+ "dataType" : "Float16",
262
+ "isOptional" : "0",
263
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
264
+ "shortDescription" : "",
265
+ "shape" : "[4, 8, 1024, 128]",
266
+ "name" : "key_cache",
267
+ "type" : "State"
268
+ },
269
+ {
270
+ "dataType" : "Float16",
271
+ "isOptional" : "0",
272
+ "formattedType" : "State (Float16 4 × 8 × 1024 × 128)",
273
+ "shortDescription" : "",
274
+ "shape" : "[4, 8, 1024, 128]",
275
+ "name" : "value_cache",
276
+ "type" : "State"
277
+ }
278
+ ],
279
+ "availability" : {
280
+ "macOS" : "15.0",
281
+ "tvOS" : "18.0",
282
+ "visionOS" : "2.0",
283
+ "watchOS" : "11.0",
284
+ "iOS" : "18.0",
285
+ "macCatalyst" : "18.0"
286
+ },
287
+ "computePrecision" : "Mixed (Float16, Int32)",
288
+ "modelType" : {
289
+ "name" : "MLModelType_mlProgram"
290
+ },
291
+ "inputSchema" : [
292
+ {
293
+ "hasShapeFlexibility" : "0",
294
+ "isOptional" : "0",
295
+ "dataType" : "Float16",
296
+ "formattedType" : "MultiArray (Float16 1 × 1 × 2560)",
297
+ "shortDescription" : "",
298
+ "shape" : "[1, 1, 2560]",
299
+ "name" : "input_embedding",
300
+ "type" : "MultiArray"
301
+ },
302
+ {
303
+ "hasShapeFlexibility" : "0",
304
+ "isOptional" : "0",
305
+ "dataType" : "Float16",
306
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
307
+ "shortDescription" : "",
308
+ "shape" : "[1, 1, 128]",
309
+ "name" : "cos",
310
+ "type" : "MultiArray"
311
+ },
312
+ {
313
+ "hasShapeFlexibility" : "0",
314
+ "isOptional" : "0",
315
+ "dataType" : "Float16",
316
+ "formattedType" : "MultiArray (Float16 1 × 1 × 128)",
317
+ "shortDescription" : "",
318
+ "shape" : "[1, 1, 128]",
319
+ "name" : "sin",
320
+ "type" : "MultiArray"
321
+ },
322
+ {
323
+ "hasShapeFlexibility" : "0",
324
+ "isOptional" : "0",
325
+ "dataType" : "Float16",
326
+ "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1024)",
327
+ "shortDescription" : "",
328
+ "shape" : "[1, 1, 1, 1024]",
329
+ "name" : "attention_mask",
330
+ "type" : "MultiArray"
331
+ },
332
+ {
333
+ "hasShapeFlexibility" : "0",
334
+ "isOptional" : "0",
335
+ "dataType" : "Int32",
336
+ "formattedType" : "MultiArray (Int32 1)",
337
+ "shortDescription" : "",
338
+ "shape" : "[1]",
339
+ "name" : "n_past",
340
+ "type" : "MultiArray"
341
+ }
342
+ ],
343
+ "defaultFunctionName" : "ar1",
344
+ "generatedClassName" : "qwen3_4B_part8_ar1_32",
345
+ "userDefinedMetadata" : {
346
+
347
+ },
348
+ "method" : "predict"
349
+ }
350
+ ]
qwen3-4B-part8-ar1-32.mlmodelc/model.mil ADDED
The diff for this file is too large to render. See raw diff
 
qwen3-4B-part8-ar1-32.mlmodelc/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fbd7a52636a3c877fc6cd56215c7d74e13666355226f68a64df0a1397b25d085
3
+ size 1585367936
utils/added_tokens.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</think>": 151668,
3
+ "</tool_call>": 151658,
4
+ "</tool_response>": 151666,
5
+ "<think>": 151667,
6
+ "<tool_call>": 151657,
7
+ "<tool_response>": 151665,
8
+ "<|box_end|>": 151649,
9
+ "<|box_start|>": 151648,
10
+ "<|endoftext|>": 151643,
11
+ "<|file_sep|>": 151664,
12
+ "<|fim_middle|>": 151660,
13
+ "<|fim_pad|>": 151662,
14
+ "<|fim_prefix|>": 151659,
15
+ "<|fim_suffix|>": 151661,
16
+ "<|im_end|>": 151645,
17
+ "<|im_start|>": 151644,
18
+ "<|image_pad|>": 151655,
19
+ "<|object_ref_end|>": 151647,
20
+ "<|object_ref_start|>": 151646,
21
+ "<|quad_end|>": 151651,
22
+ "<|quad_start|>": 151650,
23
+ "<|repo_name|>": 151663,
24
+ "<|video_pad|>": 151656,
25
+ "<|vision_end|>": 151653,
26
+ "<|vision_pad|>": 151654,
27
+ "<|vision_start|>": 151652
28
+ }
utils/chat_template.jinja ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0].role == 'system' %}
4
+ {{- messages[0].content + '\n\n' }}
5
+ {%- endif %}
6
+ {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
7
+ {%- for tool in tools %}
8
+ {{- "\n" }}
9
+ {{- tool | tojson }}
10
+ {%- endfor %}
11
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
12
+ {%- else %}
13
+ {%- if messages[0].role == 'system' %}
14
+ {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
15
+ {%- endif %}
16
+ {%- endif %}
17
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
18
+ {%- for message in messages[::-1] %}
19
+ {%- set index = (messages|length - 1) - loop.index0 %}
20
+ {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
21
+ {%- set ns.multi_step_tool = false %}
22
+ {%- set ns.last_query_index = index %}
23
+ {%- endif %}
24
+ {%- endfor %}
25
+ {%- for message in messages %}
26
+ {%- if message.content is string %}
27
+ {%- set content = message.content %}
28
+ {%- else %}
29
+ {%- set content = '' %}
30
+ {%- endif %}
31
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
32
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
33
+ {%- elif message.role == "assistant" %}
34
+ {%- set reasoning_content = '' %}
35
+ {%- if message.reasoning_content is string %}
36
+ {%- set reasoning_content = message.reasoning_content %}
37
+ {%- else %}
38
+ {%- if '</think>' in content %}
39
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
40
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
41
+ {%- endif %}
42
+ {%- endif %}
43
+ {%- if loop.index0 > ns.last_query_index %}
44
+ {%- if loop.last or (not loop.last and reasoning_content) %}
45
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
46
+ {%- else %}
47
+ {{- '<|im_start|>' + message.role + '\n' + content }}
48
+ {%- endif %}
49
+ {%- else %}
50
+ {{- '<|im_start|>' + message.role + '\n' + content }}
51
+ {%- endif %}
52
+ {%- if message.tool_calls %}
53
+ {%- for tool_call in message.tool_calls %}
54
+ {%- if (loop.first and content) or (not loop.first) %}
55
+ {{- '\n' }}
56
+ {%- endif %}
57
+ {%- if tool_call.function %}
58
+ {%- set tool_call = tool_call.function %}
59
+ {%- endif %}
60
+ {{- '<tool_call>\n{"name": "' }}
61
+ {{- tool_call.name }}
62
+ {{- '", "arguments": ' }}
63
+ {%- if tool_call.arguments is string %}
64
+ {{- tool_call.arguments }}
65
+ {%- else %}
66
+ {{- tool_call.arguments | tojson }}
67
+ {%- endif %}
68
+ {{- '}\n</tool_call>' }}
69
+ {%- endfor %}
70
+ {%- endif %}
71
+ {{- '<|im_end|>\n' }}
72
+ {%- elif message.role == "tool" %}
73
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
74
+ {{- '<|im_start|>user' }}
75
+ {%- endif %}
76
+ {{- '\n<tool_response>\n' }}
77
+ {{- content }}
78
+ {{- '\n</tool_response>' }}
79
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
80
+ {{- '<|im_end|>\n' }}
81
+ {%- endif %}
82
+ {%- endif %}
83
+ {%- endfor %}
84
+ {%- if add_generation_prompt %}
85
+ {{- '<|im_start|>assistant\n' }}
86
+ {%- if enable_thinking is defined and enable_thinking is false %}
87
+ {{- '<think>\n\n</think>\n\n' }}
88
+ {%- endif %}
89
+ {%- endif %}
utils/config.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151645,
9
+ "head_dim": 128,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 2560,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 9728,
14
+ "layer_types": [
15
+ "full_attention",
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention"
51
+ ],
52
+ "max_position_embeddings": 40960,
53
+ "max_window_layers": 36,
54
+ "model_type": "qwen3",
55
+ "num_attention_heads": 32,
56
+ "num_hidden_layers": 36,
57
+ "num_key_value_heads": 8,
58
+ "rms_norm_eps": 1e-06,
59
+ "rope_scaling": null,
60
+ "rope_theta": 1000000,
61
+ "sliding_window": null,
62
+ "tie_word_embeddings": true,
63
+ "torch_dtype": "bfloat16",
64
+ "transformers_version": "4.55.0",
65
+ "use_cache": true,
66
+ "use_sliding_window": false,
67
+ "vocab_size": 151936
68
+ }
utils/embed.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:008bff197aecdd25bc973e0191e100140ae6bbfc5fa7647a49abbfa0b0233b0f
3
+ size 1555825746