mlboydaisuke commited on
Commit
06cc985
·
verified ·
1 Parent(s): 7f94c79

Upload models.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. models.json +633 -239
models.json CHANGED
@@ -3,18 +3,78 @@
3
  "updated_at": "2026-04-10",
4
  "min_app_version": "1.0",
5
  "categories": [
6
- { "id": "segmentation", "name": "Segmentation", "icon": "person.and.background.dotted", "order": 1 },
7
- { "id": "enhancement", "name": "Image Enhancement", "icon": "wand.and.stars", "order": 2 },
8
- { "id": "detection", "name": "Object Detection", "icon": "viewfinder", "order": 3 },
9
- { "id": "depth", "name": "Depth & Geometry", "icon": "cube.transparent", "order": 4 },
10
- { "id": "vision_language", "name": "Vision-Language", "icon": "text.viewfinder", "order": 5 },
11
- { "id": "face", "name": "Face Processing", "icon": "face.smiling", "order": 6 },
12
- { "id": "generation", "name": "Image Generation", "icon": "sparkles", "order": 7 },
13
- { "id": "video", "name": "Video Processing", "icon": "film", "order": 8 },
14
- { "id": "audio", "name": "Audio Processing", "icon": "waveform.circle", "order": 9 },
15
- { "id": "speech", "name": "Speech & Music", "icon": "music.note", "order": 10 },
16
- { "id": "inpainting", "name": "Inpainting", "icon": "eraser", "order": 11 },
17
- { "id": "restoration", "name": "Face Restoration", "icon": "face.smiling.inverse", "order": 12 }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
18
  ],
19
  "models": [
20
  {
@@ -25,22 +85,35 @@
25
  "description_md": "High-quality background removal. Outputs foreground with alpha mask. INT8 quantized U-Net, 1024×1024 input.",
26
  "demo": {
27
  "template": "image_in_out",
28
- "config": { "input_size": 1024, "output_type": "mask" }
 
 
 
29
  },
30
  "files": [
31
  {
32
  "name": "RMBG_1_4.mlpackage.zip",
33
- "url": "TODO",
34
  "archive": "zip",
35
- "size_bytes": 50000000,
36
- "sha256": "TODO",
37
  "compute_units": "cpuAndGPU",
38
  "kind": "model"
39
  }
40
  ],
41
- "requirements": { "min_ios": "17.0", "min_ram_mb": 300 },
42
- "license": { "name": "Apache-2.0", "url": "https://huggingface.co/briaai/RMBG-1.4" },
43
- "upstream": { "name": "briaai/RMBG-1.4", "url": "https://huggingface.co/briaai/RMBG-1.4", "year": 2023 }
 
 
 
 
 
 
 
 
 
 
44
  },
45
  {
46
  "id": "ddcolor",
@@ -50,22 +123,35 @@
50
  "description_md": "Automatic grayscale image colorization via dual decoders. 512×512 input, processes in LAB color space.",
51
  "demo": {
52
  "template": "image_in_out",
53
- "config": { "input_size": 512, "output_type": "image" }
 
 
 
54
  },
55
  "files": [
56
  {
57
- "name": "DDColor.mlpackage.zip",
58
- "url": "TODO",
59
  "archive": "zip",
60
- "size_bytes": 35000000,
61
- "sha256": "TODO",
62
  "compute_units": "all",
63
  "kind": "model"
64
  }
65
  ],
66
- "requirements": { "min_ios": "17.0", "min_ram_mb": 400 },
67
- "license": { "name": "Apache-2.0", "url": "https://github.com/piddnad/DDColor" },
68
- "upstream": { "name": "piddnad/DDColor", "url": "https://github.com/piddnad/DDColor", "year": 2023 }
 
 
 
 
 
 
 
 
 
 
69
  },
70
  {
71
  "id": "sinsr",
@@ -75,40 +161,53 @@
75
  "description_md": "4× super-resolution via single-step diffusion. 256×256 input → 1024×1024 output. Swin Transformer denoiser (FP32 required).",
76
  "demo": {
77
  "template": "image_in_out",
78
- "config": { "input_size": 256, "output_type": "image" }
 
 
 
79
  },
80
  "files": [
81
  {
82
  "name": "SinSR_Encoder.mlpackage.zip",
83
- "url": "TODO",
84
  "archive": "zip",
85
- "size_bytes": 40000000,
86
- "sha256": "TODO",
87
  "compute_units": "cpuAndGPU",
88
  "kind": "model"
89
  },
90
  {
91
  "name": "SinSR_Denoiser.mlpackage.zip",
92
- "url": "TODO",
93
  "archive": "zip",
94
- "size_bytes": 440000000,
95
- "sha256": "TODO",
96
  "compute_units": "cpuOnly",
97
  "kind": "model"
98
  },
99
  {
100
  "name": "SinSR_Decoder.mlpackage.zip",
101
- "url": "TODO",
102
  "archive": "zip",
103
- "size_bytes": 60000000,
104
- "sha256": "TODO",
105
  "compute_units": "cpuAndGPU",
106
  "kind": "model"
107
  }
108
  ],
109
- "requirements": { "min_ios": "17.0", "min_ram_mb": 600 },
110
- "license": { "name": "Apache-2.0", "url": "https://github.com/wyf0912/SinSR" },
111
- "upstream": { "name": "wyf0912/SinSR", "url": "https://github.com/wyf0912/SinSR", "year": 2024 }
 
 
 
 
 
 
 
 
 
 
112
  },
113
  {
114
  "id": "efficientad",
@@ -118,7 +217,10 @@
118
  "description_md": "Lightweight unsupervised anomaly detection. 256×256 input → anomaly heatmap + score. Industrial quality inspection.",
119
  "demo": {
120
  "template": "image_in_out",
121
- "config": { "input_size": 256, "output_type": "image" }
 
 
 
122
  },
123
  "files": [
124
  {
@@ -131,9 +233,19 @@
131
  "kind": "model"
132
  }
133
  ],
134
- "requirements": { "min_ios": "17.0", "min_ram_mb": 200 },
135
- "license": { "name": "MIT", "url": "https://github.com/nelson1425/EfficientAD" },
136
- "upstream": { "name": "nelson1425/EfficientAD", "url": "https://github.com/nelson1425/EfficientAD", "year": 2023 }
 
 
 
 
 
 
 
 
 
 
137
  },
138
  {
139
  "id": "yolo26s",
@@ -143,22 +255,35 @@
143
  "description_md": "NMS-free object detection. 640×640 input, output [1,300,6]: x1,y1,x2,y2,confidence,class_id. 80 COCO classes.",
144
  "demo": {
145
  "template": "image_detection",
146
- "config": { "input_size": 640, "confidence_threshold": 0.25 }
 
 
 
147
  },
148
  "files": [
149
  {
150
  "name": "yolo26s.mlpackage.zip",
151
- "url": "TODO",
152
  "archive": "zip",
153
- "size_bytes": 18000000,
154
- "sha256": "TODO",
155
  "compute_units": "all",
156
  "kind": "model"
157
  }
158
  ],
159
- "requirements": { "min_ios": "17.0", "min_ram_mb": 300 },
160
- "license": { "name": "AGPL-3.0", "url": "https://github.com/ultralytics/ultralytics" },
161
- "upstream": { "name": "ultralytics/ultralytics", "url": "https://github.com/ultralytics/ultralytics", "year": 2026 }
 
 
 
 
 
 
 
 
 
 
162
  },
163
  {
164
  "id": "yolov9s",
@@ -168,22 +293,35 @@
168
  "description_md": "YOLOv9 small with Vision framework NMS. 640×640 input. PGI + GELAN architecture.",
169
  "demo": {
170
  "template": "image_detection",
171
- "config": { "input_size": 640, "confidence_threshold": 0.25 }
 
 
 
172
  },
173
  "files": [
174
  {
175
- "name": "yolov9s.mlpackage.zip",
176
- "url": "TODO",
177
  "archive": "zip",
178
- "size_bytes": 14000000,
179
- "sha256": "TODO",
180
  "compute_units": "all",
181
  "kind": "model"
182
  }
183
  ],
184
- "requirements": { "min_ios": "17.0", "min_ram_mb": 300 },
185
- "license": { "name": "AGPL-3.0", "url": "https://github.com/WongKinYiu/yolov9" },
186
- "upstream": { "name": "WongKinYiu/yolov9", "url": "https://github.com/WongKinYiu/yolov9", "year": 2024 }
 
 
 
 
 
 
 
 
 
 
187
  },
188
  {
189
  "id": "yolov10n",
@@ -193,22 +331,35 @@
193
  "description_md": "YOLOv10 nano with Vision framework NMS. 640×640 input. Dual-assignment strategy.",
194
  "demo": {
195
  "template": "image_detection",
196
- "config": { "input_size": 640, "confidence_threshold": 0.25 }
 
 
 
197
  },
198
  "files": [
199
  {
200
- "name": "yolov10n.mlpackage.zip",
201
- "url": "TODO",
202
  "archive": "zip",
203
- "size_bytes": 14000000,
204
- "sha256": "TODO",
205
  "compute_units": "all",
206
  "kind": "model"
207
  }
208
  ],
209
- "requirements": { "min_ios": "17.0", "min_ram_mb": 300 },
210
- "license": { "name": "AGPL-3.0", "url": "https://github.com/THU-MIG/yolov10" },
211
- "upstream": { "name": "THU-MIG/yolov10", "url": "https://github.com/THU-MIG/yolov10", "year": 2024 }
 
 
 
 
 
 
 
 
 
 
212
  },
213
  {
214
  "id": "moge2_vitb_normal_504",
@@ -220,24 +371,39 @@
220
  "template": "depth_visualization",
221
  "config": {
222
  "input_size": 504,
223
- "output_keys": ["depth", "normal", "mask", "metric_scale"],
 
 
 
 
 
224
  "depth_unit": "meters"
225
  }
226
  },
227
  "files": [
228
  {
229
  "name": "MoGe2_ViTB_Normal_504.mlpackage.zip",
230
- "url": "https://github.com/john-rocky/CoreML-Models/releases/download/moge2-v1/MoGe2_ViTB_Normal_504.mlpackage.zip",
231
  "archive": "zip",
232
- "size_bytes": 209715200,
233
- "sha256": "TODO",
234
  "compute_units": "all",
235
  "kind": "model"
236
  }
237
  ],
238
- "requirements": { "min_ios": "17.0", "min_ram_mb": 600 },
239
- "license": { "name": "MIT", "url": "https://github.com/microsoft/MoGe/blob/main/LICENSE" },
240
- "upstream": { "name": "microsoft/MoGe", "url": "https://github.com/microsoft/MoGe", "year": 2025 }
 
 
 
 
 
 
 
 
 
 
241
  },
242
  {
243
  "id": "siglip",
@@ -259,33 +425,43 @@
259
  "files": [
260
  {
261
  "name": "SigLIP_ImageEncoder.mlpackage.zip",
262
- "url": "TODO",
263
  "archive": "zip",
264
- "size_bytes": 350000000,
265
- "sha256": "TODO",
266
  "compute_units": "cpuOnly",
267
  "kind": "model"
268
  },
269
  {
270
  "name": "SigLIP_TextEncoder.mlpackage.zip",
271
- "url": "TODO",
272
  "archive": "zip",
273
- "size_bytes": 350000000,
274
- "sha256": "TODO",
275
  "compute_units": "cpuOnly",
276
  "kind": "model"
277
  },
278
  {
279
  "name": "siglip_vocab.json",
280
- "url": "TODO",
281
- "size_bytes": 1000000,
282
- "sha256": "TODO",
283
  "kind": "vocab"
284
  }
285
  ],
286
- "requirements": { "min_ios": "17.0", "min_ram_mb": 800 },
287
- "license": { "name": "Apache-2.0", "url": "https://github.com/google-research/big_vision" },
288
- "upstream": { "name": "google-research/big_vision", "url": "https://github.com/google-research/big_vision", "year": 2023 }
 
 
 
 
 
 
 
 
 
 
289
  },
290
  {
291
  "id": "florence2",
@@ -303,51 +479,88 @@
303
  "decoder": "Florence2Decoder.mlpackage.zip",
304
  "vocab_file": "florence2_vocab.json",
305
  "tasks": {
306
- "caption": [0, 2264, 473, 5, 2274, 6190, 116, 2],
307
- "detailed_caption": [0, 2264, 473, 5, 31962, 2274, 6190, 116, 2],
308
- "ocr": [0, 2264, 473, 5, 71307, 116, 2]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
309
  }
310
  }
311
  },
312
  "files": [
313
  {
314
  "name": "Florence2VisionEncoder.mlpackage.zip",
315
- "url": "TODO",
316
  "archive": "zip",
317
- "size_bytes": 400000000,
318
- "sha256": "TODO",
319
  "compute_units": "cpuOnly",
320
  "kind": "model"
321
  },
322
  {
323
  "name": "Florence2TextEncoder.mlpackage.zip",
324
- "url": "TODO",
325
  "archive": "zip",
326
- "size_bytes": 450000000,
327
- "sha256": "TODO",
328
  "compute_units": "cpuOnly",
329
  "kind": "model"
330
  },
331
  {
332
  "name": "Florence2Decoder.mlpackage.zip",
333
- "url": "TODO",
334
  "archive": "zip",
335
- "size_bytes": 1400000000,
336
- "sha256": "TODO",
337
  "compute_units": "cpuOnly",
338
  "kind": "model"
339
  },
340
  {
341
  "name": "florence2_vocab.json",
342
- "url": "TODO",
343
- "size_bytes": 500000,
344
- "sha256": "TODO",
345
  "kind": "vocab"
346
  }
347
  ],
348
- "requirements": { "min_ios": "17.0", "min_ram_mb": 1200 },
349
- "license": { "name": "MIT", "url": "https://huggingface.co/microsoft/Florence-2-base" },
350
- "upstream": { "name": "microsoft/Florence-2", "url": "https://huggingface.co/microsoft/Florence-2-base", "year": 2024 }
 
 
 
 
 
 
 
 
 
 
351
  },
352
  {
353
  "id": "adaface",
@@ -357,7 +570,11 @@
357
  "description_md": "Face recognition via 512-dim embeddings. IR-18 backbone, 112×112 face crop input. Compare faces by cosine similarity.",
358
  "demo": {
359
  "template": "face_compare",
360
- "config": { "input_size": 112, "embedding_dim": 512, "match_threshold": 0.6 }
 
 
 
 
361
  },
362
  "files": [
363
  {
@@ -370,9 +587,19 @@
370
  "kind": "model"
371
  }
372
  ],
373
- "requirements": { "min_ios": "17.0", "min_ram_mb": 200 },
374
- "license": { "name": "MIT", "url": "https://github.com/mk-minchul/AdaFace" },
375
- "upstream": { "name": "mk-minchul/AdaFace", "url": "https://github.com/mk-minchul/AdaFace", "year": 2022 }
 
 
 
 
 
 
 
 
 
 
376
  },
377
  {
378
  "id": "hypersd",
@@ -399,58 +626,68 @@
399
  "files": [
400
  {
401
  "name": "HyperSDTextEncoder.mlpackage.zip",
402
- "url": "https://github.com/john-rocky/CoreML-Models/releases/download/hypersd-v1/HyperSDTextEncoder.mlpackage.zip",
403
  "archive": "zip",
404
- "size_bytes": 235000000,
405
- "sha256": "TODO",
406
  "compute_units": "cpuAndNeuralEngine",
407
  "kind": "model"
408
  },
409
  {
410
  "name": "HyperSDUnetChunk1.mlpackage.zip",
411
- "url": "https://github.com/john-rocky/CoreML-Models/releases/download/hypersd-v1/HyperSDUnetChunk1.mlpackage.zip",
412
  "archive": "zip",
413
- "size_bytes": 318000000,
414
- "sha256": "TODO",
415
  "compute_units": "cpuAndNeuralEngine",
416
  "kind": "model"
417
  },
418
  {
419
  "name": "HyperSDUnetChunk2.mlpackage.zip",
420
- "url": "https://github.com/john-rocky/CoreML-Models/releases/download/hypersd-v1/HyperSDUnetChunk2.mlpackage.zip",
421
  "archive": "zip",
422
- "size_bytes": 299000000,
423
- "sha256": "TODO",
424
  "compute_units": "cpuAndNeuralEngine",
425
  "kind": "model"
426
  },
427
  {
428
  "name": "HyperSDVAEDecoder.mlpackage.zip",
429
- "url": "https://github.com/john-rocky/CoreML-Models/releases/download/hypersd-v1/HyperSDVAEDecoder.mlpackage.zip",
430
  "archive": "zip",
431
- "size_bytes": 95000000,
432
- "sha256": "TODO",
433
  "compute_units": "cpuAndNeuralEngine",
434
  "kind": "model"
435
  },
436
  {
437
  "name": "vocab.json",
438
- "url": "https://github.com/john-rocky/CoreML-Models/releases/download/hypersd-v1/vocab.json",
439
- "size_bytes": 1600000,
440
- "sha256": "TODO",
441
  "kind": "vocab"
442
  },
443
  {
444
  "name": "merges.txt",
445
- "url": "https://github.com/john-rocky/CoreML-Models/releases/download/hypersd-v1/merges.txt",
446
- "size_bytes": 525000,
447
- "sha256": "TODO",
448
  "kind": "vocab"
449
  }
450
  ],
451
- "requirements": { "min_ios": "17.0", "min_ram_mb": 1000 },
452
- "license": { "name": "OpenRAIL-M", "url": "https://huggingface.co/ByteDance/Hyper-SD" },
453
- "upstream": { "name": "ByteDance/Hyper-SD", "url": "https://huggingface.co/ByteDance/Hyper-SD", "year": 2024 }
 
 
 
 
 
 
 
 
 
 
454
  },
455
  {
456
  "id": "matanyone",
@@ -471,54 +708,64 @@
471
  },
472
  "files": [
473
  {
474
- "name": "MatAnyone_Encoder.mlpackage.zip",
475
- "url": "TODO",
476
  "archive": "zip",
477
- "size_bytes": 20000000,
478
- "sha256": "TODO",
479
  "compute_units": "all",
480
  "kind": "model"
481
  },
482
  {
483
- "name": "MatAnyone_MaskEncoder.mlpackage.zip",
484
- "url": "TODO",
485
  "archive": "zip",
486
- "size_bytes": 10000000,
487
- "sha256": "TODO",
488
  "compute_units": "all",
489
  "kind": "model"
490
  },
491
  {
492
- "name": "MatAnyone_ReadFirst.mlpackage.zip",
493
- "url": "TODO",
494
  "archive": "zip",
495
- "size_bytes": 15000000,
496
- "sha256": "TODO",
497
  "compute_units": "cpuOnly",
498
  "kind": "model"
499
  },
500
  {
501
- "name": "MatAnyone_Read.mlpackage.zip",
502
- "url": "TODO",
503
  "archive": "zip",
504
- "size_bytes": 20000000,
505
- "sha256": "TODO",
506
  "compute_units": "cpuOnly",
507
  "kind": "model"
508
  },
509
  {
510
- "name": "MatAnyone_Decoder.mlpackage.zip",
511
- "url": "TODO",
512
  "archive": "zip",
513
- "size_bytes": 35000000,
514
- "sha256": "TODO",
515
  "compute_units": "all",
516
  "kind": "model"
517
  }
518
  ],
519
- "requirements": { "min_ios": "17.0", "min_ram_mb": 800 },
520
- "license": { "name": "MIT", "url": "https://github.com/pq-yang/MatAnyone" },
521
- "upstream": { "name": "pq-yang/MatAnyone", "url": "https://github.com/pq-yang/MatAnyone", "year": 2025 }
 
 
 
 
 
 
 
 
 
 
522
  },
523
  {
524
  "id": "demucs",
@@ -531,23 +778,38 @@
531
  "config": {
532
  "sample_rate": 44100,
533
  "segment_length": 343980,
534
- "output_stems": ["drums", "bass", "vocals", "other"]
 
 
 
 
 
535
  }
536
  },
537
  "files": [
538
  {
539
  "name": "HTDemucs_SourceSeparation_F32.mlpackage.zip",
540
- "url": "TODO",
541
  "archive": "zip",
542
- "size_bytes": 360000000,
543
- "sha256": "TODO",
544
  "compute_units": "cpuOnly",
545
  "kind": "model"
546
  }
547
  ],
548
- "requirements": { "min_ios": "17.0", "min_ram_mb": 1000 },
549
- "license": { "name": "MIT", "url": "https://github.com/adefossez/demucs" },
550
- "upstream": { "name": "adefossez/demucs", "url": "https://github.com/adefossez/demucs", "year": 2021 }
 
 
 
 
 
 
 
 
 
 
551
  },
552
  {
553
  "id": "kokoro",
@@ -561,57 +823,73 @@
561
  "mode": "tts",
562
  "sample_rate": 24000,
563
  "vocab_file": "kokoro_vocab.json",
564
- "voices": ["af_heart", "af_bella", "am_michael", "bf_emma", "bm_george"]
 
 
 
 
 
 
565
  }
566
  },
567
  "files": [
568
  {
569
  "name": "Kokoro_Predictor.mlpackage.zip",
570
- "url": "TODO",
571
  "archive": "zip",
572
- "size_bytes": 75000000,
573
- "sha256": "TODO",
574
  "compute_units": "cpuAndGPU",
575
  "kind": "model"
576
  },
577
  {
578
  "name": "Kokoro_Decoder_128.mlpackage.zip",
579
- "url": "TODO",
580
  "archive": "zip",
581
- "size_bytes": 238000000,
582
- "sha256": "TODO",
583
  "compute_units": "all",
584
  "kind": "model"
585
  },
586
  {
587
  "name": "Kokoro_Decoder_256.mlpackage.zip",
588
- "url": "TODO",
589
  "archive": "zip",
590
- "size_bytes": 241000000,
591
- "sha256": "TODO",
592
  "compute_units": "all",
593
  "kind": "model"
594
  },
595
  {
596
  "name": "Kokoro_Decoder_512.mlpackage.zip",
597
- "url": "TODO",
598
  "archive": "zip",
599
- "size_bytes": 246000000,
600
- "sha256": "TODO",
601
  "compute_units": "all",
602
  "kind": "model"
603
  },
604
  {
605
  "name": "kokoro_vocab.json",
606
- "url": "TODO",
607
- "size_bytes": 5000,
608
- "sha256": "TODO",
609
  "kind": "vocab"
610
  }
611
  ],
612
- "requirements": { "min_ios": "17.0", "min_ram_mb": 1000 },
613
- "license": { "name": "Apache-2.0", "url": "https://huggingface.co/hexgrad/Kokoro-82M" },
614
- "upstream": { "name": "hexgrad/Kokoro-82M", "url": "https://huggingface.co/hexgrad/Kokoro-82M", "year": 2024 }
 
 
 
 
 
 
 
 
 
 
615
  },
616
  {
617
  "id": "stable_audio",
@@ -630,51 +908,61 @@
630
  "files": [
631
  {
632
  "name": "StableAudioT5Encoder.mlpackage.zip",
633
- "url": "TODO",
634
  "archive": "zip",
635
- "size_bytes": 105000000,
636
- "sha256": "TODO",
637
  "compute_units": "cpuOnly",
638
  "kind": "model"
639
  },
640
  {
641
  "name": "StableAudioNumberEmbedder.mlpackage.zip",
642
- "url": "TODO",
643
  "archive": "zip",
644
- "size_bytes": 400000,
645
- "sha256": "TODO",
646
  "compute_units": "cpuOnly",
647
  "kind": "model"
648
  },
649
  {
650
  "name": "StableAudioDiT.mlpackage.zip",
651
- "url": "TODO",
652
  "archive": "zip",
653
- "size_bytes": 326000000,
654
- "sha256": "TODO",
655
  "compute_units": "cpuOnly",
656
  "kind": "model"
657
  },
658
  {
659
  "name": "StableAudioVAEDecoder.mlpackage.zip",
660
- "url": "TODO",
661
  "archive": "zip",
662
- "size_bytes": 149000000,
663
- "sha256": "TODO",
664
  "compute_units": "cpuAndGPU",
665
  "kind": "model"
666
  },
667
  {
668
  "name": "t5_vocab.json",
669
- "url": "TODO",
670
- "size_bytes": 800000,
671
- "sha256": "TODO",
672
  "kind": "vocab"
673
  }
674
  ],
675
- "requirements": { "min_ios": "17.0", "min_ram_mb": 1200 },
676
- "license": { "name": "custom", "url": "https://huggingface.co/stabilityai/stable-audio-open-small" },
677
- "upstream": { "name": "stabilityai/stable-audio-open-small", "url": "https://huggingface.co/stabilityai/stable-audio-open-small", "year": 2024 }
 
 
 
 
 
 
 
 
 
 
678
  },
679
  {
680
  "id": "basicpitch",
@@ -704,9 +992,19 @@
704
  "kind": "model"
705
  }
706
  ],
707
- "requirements": { "min_ios": "17.0", "min_ram_mb": 200 },
708
- "license": { "name": "Apache-2.0", "url": "https://github.com/spotify/basic-pitch" },
709
- "upstream": { "name": "spotify/basic-pitch", "url": "https://github.com/spotify/basic-pitch", "year": 2022 }
 
 
 
 
 
 
 
 
 
 
710
  },
711
  {
712
  "id": "diarization",
@@ -718,23 +1016,35 @@
718
  "template": "audio_in_out",
719
  "config": {
720
  "sample_rate": 16000,
721
- "output_stems": ["speaker_timeline"]
 
 
722
  }
723
  },
724
  "files": [
725
  {
726
- "name": "Pyannote_Segmentation3_0.mlpackage.zip",
727
- "url": "TODO",
728
  "archive": "zip",
729
- "size_bytes": 25000000,
730
- "sha256": "TODO",
731
  "compute_units": "cpuAndGPU",
732
  "kind": "model"
733
  }
734
  ],
735
- "requirements": { "min_ios": "17.0", "min_ram_mb": 200 },
736
- "license": { "name": "MIT", "url": "https://github.com/pyannote/pyannote-audio" },
737
- "upstream": { "name": "pyannote/pyannote-audio", "url": "https://github.com/pyannote/pyannote-audio", "year": 2021 }
 
 
 
 
 
 
 
 
 
 
738
  },
739
  {
740
  "id": "openvoice",
@@ -746,32 +1056,44 @@
746
  "template": "audio_in_out",
747
  "config": {
748
  "sample_rate": 22050,
749
- "output_stems": ["converted"]
 
 
750
  }
751
  },
752
  "files": [
753
  {
754
  "name": "OpenVoice_SpeakerEncoder.mlpackage.zip",
755
- "url": "TODO",
756
  "archive": "zip",
757
- "size_bytes": 35000000,
758
- "sha256": "TODO",
759
  "compute_units": "cpuAndGPU",
760
  "kind": "model"
761
  },
762
  {
763
  "name": "OpenVoice_VoiceConverter.mlpackage.zip",
764
- "url": "TODO",
765
  "archive": "zip",
766
- "size_bytes": 100000000,
767
- "sha256": "TODO",
768
  "compute_units": "cpuAndGPU",
769
  "kind": "model"
770
  }
771
  ],
772
- "requirements": { "min_ios": "17.0", "min_ram_mb": 500 },
773
- "license": { "name": "MIT", "url": "https://github.com/myshell-ai/OpenVoice" },
774
- "upstream": { "name": "myshell-ai/OpenVoice", "url": "https://github.com/myshell-ai/OpenVoice", "year": 2023 }
 
 
 
 
 
 
 
 
 
 
775
  },
776
  {
777
  "id": "realesrgan",
@@ -781,7 +1103,10 @@
781
  "description_md": "Real-world blind super-resolution. 4× upscale from any input. Handles noise, blur, and JPEG artifacts. 512×512 input → 2048×2048 output.",
782
  "demo": {
783
  "template": "image_in_out",
784
- "config": { "input_size": 512, "output_type": "image" }
 
 
 
785
  },
786
  "files": [
787
  {
@@ -794,9 +1119,19 @@
794
  "kind": "model"
795
  }
796
  ],
797
- "requirements": { "min_ios": "17.0", "min_ram_mb": 500 },
798
- "license": { "name": "BSD-3-Clause", "url": "https://github.com/xinntao/Real-ESRGAN/blob/master/LICENSE" },
799
- "upstream": { "name": "xinntao/Real-ESRGAN", "url": "https://github.com/xinntao/Real-ESRGAN", "year": 2021 }
 
 
 
 
 
 
 
 
 
 
800
  },
801
  {
802
  "id": "gfpgan",
@@ -806,7 +1141,10 @@
806
  "description_md": "Blind face restoration with generative facial prior. Restores degraded face photos to high quality. 512×512 input/output.",
807
  "demo": {
808
  "template": "image_in_out",
809
- "config": { "input_size": 512, "output_type": "image" }
 
 
 
810
  },
811
  "files": [
812
  {
@@ -819,9 +1157,19 @@
819
  "kind": "model"
820
  }
821
  ],
822
- "requirements": { "min_ios": "17.0", "min_ram_mb": 600 },
823
- "license": { "name": "Apache-2.0", "url": "https://github.com/TencentARC/GFPGAN/blob/master/LICENSE" },
824
- "upstream": { "name": "TencentARC/GFPGAN", "url": "https://github.com/TencentARC/GFPGAN", "year": 2021 }
 
 
 
 
 
 
 
 
 
 
825
  },
826
  {
827
  "id": "rfdetr_n",
@@ -850,9 +1198,19 @@
850
  "kind": "model"
851
  }
852
  ],
853
- "requirements": { "min_ios": "17.0", "min_ram_mb": 400 },
854
- "license": { "name": "Apache-2.0", "url": "https://github.com/roboflow/rf-detr/blob/main/LICENSE" },
855
- "upstream": { "name": "roboflow/rf-detr", "url": "https://github.com/roboflow/rf-detr", "year": 2025 }
 
 
 
 
 
 
 
 
 
 
856
  },
857
  {
858
  "id": "face_parsing",
@@ -862,7 +1220,11 @@
862
  "description_md": "Semantic face parsing into 19 regions: skin, nose, eyes, eyebrows, ears, mouth, lip, hair, hat, eyeglass, earring, necklace, neck, cloth, background. 512×512 input.",
863
  "demo": {
864
  "template": "image_in_out",
865
- "config": { "input_size": 512, "output_type": "segmap", "num_classes": 19 }
 
 
 
 
866
  },
867
  "files": [
868
  {
@@ -875,9 +1237,19 @@
875
  "kind": "model"
876
  }
877
  ],
878
- "requirements": { "min_ios": "17.0", "min_ram_mb": 300 },
879
- "license": { "name": "MIT", "url": "https://github.com/zllrunning/face-parsing.PyTorch/blob/master/LICENSE" },
880
- "upstream": { "name": "zllrunning/face-parsing.PyTorch", "url": "https://github.com/zllrunning/face-parsing.PyTorch", "year": 2019 }
 
 
 
 
 
 
 
 
 
 
881
  },
882
  {
883
  "id": "mobilesam",
@@ -904,9 +1276,19 @@
904
  "kind": "model"
905
  }
906
  ],
907
- "requirements": { "min_ios": "17.0", "min_ram_mb": 300 },
908
- "license": { "name": "Apache-2.0", "url": "https://github.com/ChaoningZhang/MobileSAM/blob/master/LICENSE" },
909
- "upstream": { "name": "ChaoningZhang/MobileSAM", "url": "https://github.com/ChaoningZhang/MobileSAM", "year": 2023 }
 
 
 
 
 
 
 
 
 
 
910
  },
911
  {
912
  "id": "lama",
@@ -916,7 +1298,9 @@
916
  "description_md": "Resolution-robust large mask inpainting. Draw over unwanted objects to remove them. Fast Fourier convolutions for global context. 800×800 input.",
917
  "demo": {
918
  "template": "inpainting",
919
- "config": { "input_size": 800 }
 
 
920
  },
921
  "files": [
922
  {
@@ -929,9 +1313,19 @@
929
  "kind": "model"
930
  }
931
  ],
932
- "requirements": { "min_ios": "17.0", "min_ram_mb": 600 },
933
- "license": { "name": "Apache-2.0", "url": "https://github.com/advimman/lama/blob/main/LICENSE" },
934
- "upstream": { "name": "advimman/lama", "url": "https://github.com/advimman/lama", "year": 2022 }
 
 
 
 
 
 
 
 
 
 
935
  }
936
  ]
937
  }
 
3
  "updated_at": "2026-04-10",
4
  "min_app_version": "1.0",
5
  "categories": [
6
+ {
7
+ "id": "segmentation",
8
+ "name": "Segmentation",
9
+ "icon": "person.and.background.dotted",
10
+ "order": 1
11
+ },
12
+ {
13
+ "id": "enhancement",
14
+ "name": "Image Enhancement",
15
+ "icon": "wand.and.stars",
16
+ "order": 2
17
+ },
18
+ {
19
+ "id": "detection",
20
+ "name": "Object Detection",
21
+ "icon": "viewfinder",
22
+ "order": 3
23
+ },
24
+ {
25
+ "id": "depth",
26
+ "name": "Depth & Geometry",
27
+ "icon": "cube.transparent",
28
+ "order": 4
29
+ },
30
+ {
31
+ "id": "vision_language",
32
+ "name": "Vision-Language",
33
+ "icon": "text.viewfinder",
34
+ "order": 5
35
+ },
36
+ {
37
+ "id": "face",
38
+ "name": "Face Processing",
39
+ "icon": "face.smiling",
40
+ "order": 6
41
+ },
42
+ {
43
+ "id": "generation",
44
+ "name": "Image Generation",
45
+ "icon": "sparkles",
46
+ "order": 7
47
+ },
48
+ {
49
+ "id": "video",
50
+ "name": "Video Processing",
51
+ "icon": "film",
52
+ "order": 8
53
+ },
54
+ {
55
+ "id": "audio",
56
+ "name": "Audio Processing",
57
+ "icon": "waveform.circle",
58
+ "order": 9
59
+ },
60
+ {
61
+ "id": "speech",
62
+ "name": "Speech & Music",
63
+ "icon": "music.note",
64
+ "order": 10
65
+ },
66
+ {
67
+ "id": "inpainting",
68
+ "name": "Inpainting",
69
+ "icon": "eraser",
70
+ "order": 11
71
+ },
72
+ {
73
+ "id": "restoration",
74
+ "name": "Face Restoration",
75
+ "icon": "face.smiling.inverse",
76
+ "order": 12
77
+ }
78
  ],
79
  "models": [
80
  {
 
85
  "description_md": "High-quality background removal. Outputs foreground with alpha mask. INT8 quantized U-Net, 1024×1024 input.",
86
  "demo": {
87
  "template": "image_in_out",
88
+ "config": {
89
+ "input_size": 1024,
90
+ "output_type": "mask"
91
+ }
92
  },
93
  "files": [
94
  {
95
  "name": "RMBG_1_4.mlpackage.zip",
96
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/rmbg/RMBG_1_4.mlpackage.zip",
97
  "archive": "zip",
98
+ "size_bytes": 38771210,
99
+ "sha256": "a80dbb5f04c922a8fa698c38592e4e52af4e62471d70bc7c59c28a3355a1da95",
100
  "compute_units": "cpuAndGPU",
101
  "kind": "model"
102
  }
103
  ],
104
+ "requirements": {
105
+ "min_ios": "17.0",
106
+ "min_ram_mb": 300
107
+ },
108
+ "license": {
109
+ "name": "Apache-2.0",
110
+ "url": "https://huggingface.co/briaai/RMBG-1.4"
111
+ },
112
+ "upstream": {
113
+ "name": "briaai/RMBG-1.4",
114
+ "url": "https://huggingface.co/briaai/RMBG-1.4",
115
+ "year": 2023
116
+ }
117
  },
118
  {
119
  "id": "ddcolor",
 
123
  "description_md": "Automatic grayscale image colorization via dual decoders. 512×512 input, processes in LAB color space.",
124
  "demo": {
125
  "template": "image_in_out",
126
+ "config": {
127
+ "input_size": 512,
128
+ "output_type": "lab_ab"
129
+ }
130
  },
131
  "files": [
132
  {
133
+ "name": "DDColor_Tiny.mlpackage.zip",
134
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/ddcolor/DDColor_Tiny.mlpackage.zip",
135
  "archive": "zip",
136
+ "size_bytes": 212344570,
137
+ "sha256": "bfecea37d66005f602efe13978360b8e4707923234c3d1d00beeb4e36cb1b02c",
138
  "compute_units": "all",
139
  "kind": "model"
140
  }
141
  ],
142
+ "requirements": {
143
+ "min_ios": "17.0",
144
+ "min_ram_mb": 400
145
+ },
146
+ "license": {
147
+ "name": "Apache-2.0",
148
+ "url": "https://github.com/piddnad/DDColor"
149
+ },
150
+ "upstream": {
151
+ "name": "piddnad/DDColor",
152
+ "url": "https://github.com/piddnad/DDColor",
153
+ "year": 2023
154
+ }
155
  },
156
  {
157
  "id": "sinsr",
 
161
  "description_md": "4× super-resolution via single-step diffusion. 256×256 input → 1024×1024 output. Swin Transformer denoiser (FP32 required).",
162
  "demo": {
163
  "template": "image_in_out",
164
+ "config": {
165
+ "input_size": 256,
166
+ "output_type": "image"
167
+ }
168
  },
169
  "files": [
170
  {
171
  "name": "SinSR_Encoder.mlpackage.zip",
172
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/sinsr/SinSR_Encoder.mlpackage.zip",
173
  "archive": "zip",
174
+ "size_bytes": 41246338,
175
+ "sha256": "fdec09d17561ec1bb5a2e829683d48c2b45e76b876285619a6e29a3523b8b7e2",
176
  "compute_units": "cpuAndGPU",
177
  "kind": "model"
178
  },
179
  {
180
  "name": "SinSR_Denoiser.mlpackage.zip",
181
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/sinsr/SinSR_Denoiser.mlpackage.zip",
182
  "archive": "zip",
183
+ "size_bytes": 440014511,
184
+ "sha256": "b31374c2d539b2cdd81499d6062c801ca00e405f5a67507cd609d14e2d6d4beb",
185
  "compute_units": "cpuOnly",
186
  "kind": "model"
187
  },
188
  {
189
  "name": "SinSR_Decoder.mlpackage.zip",
190
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/sinsr/SinSR_Decoder.mlpackage.zip",
191
  "archive": "zip",
192
+ "size_bytes": 60880285,
193
+ "sha256": "b8b9a7b52d6b240cf9fb3352b286ea83eb984fd73f5dd81c9f034f0016a5cb8c",
194
  "compute_units": "cpuAndGPU",
195
  "kind": "model"
196
  }
197
  ],
198
+ "requirements": {
199
+ "min_ios": "17.0",
200
+ "min_ram_mb": 600
201
+ },
202
+ "license": {
203
+ "name": "Apache-2.0",
204
+ "url": "https://github.com/wyf0912/SinSR"
205
+ },
206
+ "upstream": {
207
+ "name": "wyf0912/SinSR",
208
+ "url": "https://github.com/wyf0912/SinSR",
209
+ "year": 2024
210
+ }
211
  },
212
  {
213
  "id": "efficientad",
 
217
  "description_md": "Lightweight unsupervised anomaly detection. 256×256 input → anomaly heatmap + score. Industrial quality inspection.",
218
  "demo": {
219
  "template": "image_in_out",
220
+ "config": {
221
+ "input_size": 256,
222
+ "output_type": "image"
223
+ }
224
  },
225
  "files": [
226
  {
 
233
  "kind": "model"
234
  }
235
  ],
236
+ "requirements": {
237
+ "min_ios": "17.0",
238
+ "min_ram_mb": 200
239
+ },
240
+ "license": {
241
+ "name": "MIT",
242
+ "url": "https://github.com/nelson1425/EfficientAD"
243
+ },
244
+ "upstream": {
245
+ "name": "nelson1425/EfficientAD",
246
+ "url": "https://github.com/nelson1425/EfficientAD",
247
+ "year": 2023
248
+ }
249
  },
250
  {
251
  "id": "yolo26s",
 
255
  "description_md": "NMS-free object detection. 640×640 input, output [1,300,6]: x1,y1,x2,y2,confidence,class_id. 80 COCO classes.",
256
  "demo": {
257
  "template": "image_detection",
258
+ "config": {
259
+ "input_size": 640,
260
+ "confidence_threshold": 0.25
261
+ }
262
  },
263
  "files": [
264
  {
265
  "name": "yolo26s.mlpackage.zip",
266
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/yolo26/yolo26s.mlpackage.zip",
267
  "archive": "zip",
268
+ "size_bytes": 17697581,
269
+ "sha256": "0ec02fb0cf2dbd6e09601cbbc00a9734156ea4c2a52b0da23a984337074c6fd4",
270
  "compute_units": "all",
271
  "kind": "model"
272
  }
273
  ],
274
+ "requirements": {
275
+ "min_ios": "17.0",
276
+ "min_ram_mb": 300
277
+ },
278
+ "license": {
279
+ "name": "AGPL-3.0",
280
+ "url": "https://github.com/ultralytics/ultralytics"
281
+ },
282
+ "upstream": {
283
+ "name": "ultralytics/ultralytics",
284
+ "url": "https://github.com/ultralytics/ultralytics",
285
+ "year": 2026
286
+ }
287
  },
288
  {
289
  "id": "yolov9s",
 
293
  "description_md": "YOLOv9 small with Vision framework NMS. 640×640 input. PGI + GELAN architecture.",
294
  "demo": {
295
  "template": "image_detection",
296
+ "config": {
297
+ "input_size": 640,
298
+ "confidence_threshold": 0.25
299
+ }
300
  },
301
  "files": [
302
  {
303
+ "name": "yolo11s.mlpackage.zip",
304
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/yolov9/yolo11s.mlpackage.zip",
305
  "archive": "zip",
306
+ "size_bytes": 17580204,
307
+ "sha256": "79e82aacc3ad20fc1eb990df6979fae9b927d4b06f33bd20ec0e1c0dcb7d1f6b",
308
  "compute_units": "all",
309
  "kind": "model"
310
  }
311
  ],
312
+ "requirements": {
313
+ "min_ios": "17.0",
314
+ "min_ram_mb": 300
315
+ },
316
+ "license": {
317
+ "name": "AGPL-3.0",
318
+ "url": "https://github.com/WongKinYiu/yolov9"
319
+ },
320
+ "upstream": {
321
+ "name": "WongKinYiu/yolov9",
322
+ "url": "https://github.com/WongKinYiu/yolov9",
323
+ "year": 2024
324
+ }
325
  },
326
  {
327
  "id": "yolov10n",
 
331
  "description_md": "YOLOv10 nano with Vision framework NMS. 640×640 input. Dual-assignment strategy.",
332
  "demo": {
333
  "template": "image_detection",
334
+ "config": {
335
+ "input_size": 640,
336
+ "confidence_threshold": 0.25
337
+ }
338
  },
339
  "files": [
340
  {
341
+ "name": "YOLOv10N.mlpackage.zip",
342
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/yolov10/YOLOv10N.mlpackage.zip",
343
  "archive": "zip",
344
+ "size_bytes": 4309168,
345
+ "sha256": "9a687144a6b0b764f508c8f544fe46b6674629b8f09a1e99d8ca69b0be899891",
346
  "compute_units": "all",
347
  "kind": "model"
348
  }
349
  ],
350
+ "requirements": {
351
+ "min_ios": "17.0",
352
+ "min_ram_mb": 300
353
+ },
354
+ "license": {
355
+ "name": "AGPL-3.0",
356
+ "url": "https://github.com/THU-MIG/yolov10"
357
+ },
358
+ "upstream": {
359
+ "name": "THU-MIG/yolov10",
360
+ "url": "https://github.com/THU-MIG/yolov10",
361
+ "year": 2024
362
+ }
363
  },
364
  {
365
  "id": "moge2_vitb_normal_504",
 
371
  "template": "depth_visualization",
372
  "config": {
373
  "input_size": 504,
374
+ "output_keys": [
375
+ "depth",
376
+ "normal",
377
+ "mask",
378
+ "metric_scale"
379
+ ],
380
  "depth_unit": "meters"
381
  }
382
  },
383
  "files": [
384
  {
385
  "name": "MoGe2_ViTB_Normal_504.mlpackage.zip",
386
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/moge2/MoGe2_ViTB_Normal_504.mlpackage.zip",
387
  "archive": "zip",
388
+ "size_bytes": 193312088,
389
+ "sha256": "f60cfb4804707a489d99e24453188cd31ddcabb299bbf6da4507edc9cecbf9e7",
390
  "compute_units": "all",
391
  "kind": "model"
392
  }
393
  ],
394
+ "requirements": {
395
+ "min_ios": "17.0",
396
+ "min_ram_mb": 600
397
+ },
398
+ "license": {
399
+ "name": "MIT",
400
+ "url": "https://github.com/microsoft/MoGe/blob/main/LICENSE"
401
+ },
402
+ "upstream": {
403
+ "name": "microsoft/MoGe",
404
+ "url": "https://github.com/microsoft/MoGe",
405
+ "year": 2025
406
+ }
407
  },
408
  {
409
  "id": "siglip",
 
425
  "files": [
426
  {
427
  "name": "SigLIP_ImageEncoder.mlpackage.zip",
428
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/siglip/SigLIP_ImageEncoder.mlpackage.zip",
429
  "archive": "zip",
430
+ "size_bytes": 170352400,
431
+ "sha256": "98f6abf5f4aa145199f4ae22305f9c1d5929eee6b126daad84783b2b2090ee24",
432
  "compute_units": "cpuOnly",
433
  "kind": "model"
434
  },
435
  {
436
  "name": "SigLIP_TextEncoder.mlpackage.zip",
437
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/siglip/SigLIP_TextEncoder.mlpackage.zip",
438
  "archive": "zip",
439
+ "size_bytes": 203975769,
440
+ "sha256": "9dead2d58705838aef7ad83c3bf4036698c78d872ca1cdd04f2c4a6272009ccf",
441
  "compute_units": "cpuOnly",
442
  "kind": "model"
443
  },
444
  {
445
  "name": "siglip_vocab.json",
446
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/siglip/siglip_vocab.json",
447
+ "size_bytes": 673754,
448
+ "sha256": "b94b3a58e04f619936b3890804dff7c478522c07515ff748cf127c5443ee5229",
449
  "kind": "vocab"
450
  }
451
  ],
452
+ "requirements": {
453
+ "min_ios": "17.0",
454
+ "min_ram_mb": 800
455
+ },
456
+ "license": {
457
+ "name": "Apache-2.0",
458
+ "url": "https://github.com/google-research/big_vision"
459
+ },
460
+ "upstream": {
461
+ "name": "google-research/big_vision",
462
+ "url": "https://github.com/google-research/big_vision",
463
+ "year": 2023
464
+ }
465
  },
466
  {
467
  "id": "florence2",
 
479
  "decoder": "Florence2Decoder.mlpackage.zip",
480
  "vocab_file": "florence2_vocab.json",
481
  "tasks": {
482
+ "caption": [
483
+ 0,
484
+ 2264,
485
+ 473,
486
+ 5,
487
+ 2274,
488
+ 6190,
489
+ 116,
490
+ 2
491
+ ],
492
+ "detailed_caption": [
493
+ 0,
494
+ 2264,
495
+ 473,
496
+ 5,
497
+ 31962,
498
+ 2274,
499
+ 6190,
500
+ 116,
501
+ 2
502
+ ],
503
+ "ocr": [
504
+ 0,
505
+ 2264,
506
+ 473,
507
+ 5,
508
+ 71307,
509
+ 116,
510
+ 2
511
+ ]
512
  }
513
  }
514
  },
515
  "files": [
516
  {
517
  "name": "Florence2VisionEncoder.mlpackage.zip",
518
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/florence2/Florence2VisionEncoder.mlpackage.zip",
519
  "archive": "zip",
520
+ "size_bytes": 81198683,
521
+ "sha256": "9422f189c21220a0f9966eb9d780856772feb55597dcc579fc4e3c88990d0046",
522
  "compute_units": "cpuOnly",
523
  "kind": "model"
524
  },
525
  {
526
  "name": "Florence2TextEncoder.mlpackage.zip",
527
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/florence2/Florence2TextEncoder.mlpackage.zip",
528
  "archive": "zip",
529
+ "size_bytes": 72742890,
530
+ "sha256": "f985deeef0408ea8aac33ac4f5c6d9635cd9c64c98b53f85031db6e27f3bfd92",
531
  "compute_units": "cpuOnly",
532
  "kind": "model"
533
  },
534
  {
535
  "name": "Florence2Decoder.mlpackage.zip",
536
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/florence2/Florence2Decoder.mlpackage.zip",
537
  "archive": "zip",
538
+ "size_bytes": 85329746,
539
+ "sha256": "fe85a6faab5281272bcd79dabfbf87d60ba1a78dd9455e2bf71c67a134d61dc5",
540
  "compute_units": "cpuOnly",
541
  "kind": "model"
542
  },
543
  {
544
  "name": "florence2_vocab.json",
545
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/florence2/florence2_vocab.json",
546
+ "size_bytes": 999352,
547
+ "sha256": "861fee9af5520403f6dbb4940d6af6627f1481b71cdc4a870f1f61344e57e645",
548
  "kind": "vocab"
549
  }
550
  ],
551
+ "requirements": {
552
+ "min_ios": "17.0",
553
+ "min_ram_mb": 1200
554
+ },
555
+ "license": {
556
+ "name": "MIT",
557
+ "url": "https://huggingface.co/microsoft/Florence-2-base"
558
+ },
559
+ "upstream": {
560
+ "name": "microsoft/Florence-2",
561
+ "url": "https://huggingface.co/microsoft/Florence-2-base",
562
+ "year": 2024
563
+ }
564
  },
565
  {
566
  "id": "adaface",
 
570
  "description_md": "Face recognition via 512-dim embeddings. IR-18 backbone, 112×112 face crop input. Compare faces by cosine similarity.",
571
  "demo": {
572
  "template": "face_compare",
573
+ "config": {
574
+ "input_size": 112,
575
+ "embedding_dim": 512,
576
+ "match_threshold": 0.6
577
+ }
578
  },
579
  "files": [
580
  {
 
587
  "kind": "model"
588
  }
589
  ],
590
+ "requirements": {
591
+ "min_ios": "17.0",
592
+ "min_ram_mb": 200
593
+ },
594
+ "license": {
595
+ "name": "MIT",
596
+ "url": "https://github.com/mk-minchul/AdaFace"
597
+ },
598
+ "upstream": {
599
+ "name": "mk-minchul/AdaFace",
600
+ "url": "https://github.com/mk-minchul/AdaFace",
601
+ "year": 2022
602
+ }
603
  },
604
  {
605
  "id": "hypersd",
 
626
  "files": [
627
  {
628
  "name": "HyperSDTextEncoder.mlpackage.zip",
629
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/hypersd/HyperSDTextEncoder.mlpackage.zip",
630
  "archive": "zip",
631
+ "size_bytes": 226397794,
632
+ "sha256": "201b0fcc3573811aac6a4e8545c695bc4fb2f7710ea0d60c227919d87b37687e",
633
  "compute_units": "cpuAndNeuralEngine",
634
  "kind": "model"
635
  },
636
  {
637
  "name": "HyperSDUnetChunk1.mlpackage.zip",
638
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/hypersd/HyperSDUnetChunk1.mlpackage.zip",
639
  "archive": "zip",
640
+ "size_bytes": 324819653,
641
+ "sha256": "279da11b8231aeeb9045f6ceabebb3a68c20a1b86ecc81aa6914b77ce76d5203",
642
  "compute_units": "cpuAndNeuralEngine",
643
  "kind": "model"
644
  },
645
  {
646
  "name": "HyperSDUnetChunk2.mlpackage.zip",
647
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/hypersd/HyperSDUnetChunk2.mlpackage.zip",
648
  "archive": "zip",
649
+ "size_bytes": 304530429,
650
+ "sha256": "0a700d11a105da589bb3e5666e38b9c72fa283149951b253fc11722e70e72faa",
651
  "compute_units": "cpuAndNeuralEngine",
652
  "kind": "model"
653
  },
654
  {
655
  "name": "HyperSDVAEDecoder.mlpackage.zip",
656
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/hypersd/HyperSDVAEDecoder.mlpackage.zip",
657
  "archive": "zip",
658
+ "size_bytes": 91282754,
659
+ "sha256": "1260371542d845a2261ed2de36c5fe3e9ccb740a6ceb59b1990705d125e8cf66",
660
  "compute_units": "cpuAndNeuralEngine",
661
  "kind": "model"
662
  },
663
  {
664
  "name": "vocab.json",
665
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/hypersd/vocab.json",
666
+ "size_bytes": 1059962,
667
+ "sha256": "e089ad92ba36837a0d31433e555c8f45fe601ab5c221d4f607ded32d9f7a4349",
668
  "kind": "vocab"
669
  },
670
  {
671
  "name": "merges.txt",
672
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/hypersd/merges.txt",
673
+ "size_bytes": 524619,
674
+ "sha256": "9fd691f7c8039210e0fced15865466c65820d09b63988b0174bfe25de299051a",
675
  "kind": "vocab"
676
  }
677
  ],
678
+ "requirements": {
679
+ "min_ios": "17.0",
680
+ "min_ram_mb": 1000
681
+ },
682
+ "license": {
683
+ "name": "OpenRAIL-M",
684
+ "url": "https://huggingface.co/ByteDance/Hyper-SD"
685
+ },
686
+ "upstream": {
687
+ "name": "ByteDance/Hyper-SD",
688
+ "url": "https://huggingface.co/ByteDance/Hyper-SD",
689
+ "year": 2024
690
+ }
691
  },
692
  {
693
  "id": "matanyone",
 
708
  },
709
  "files": [
710
  {
711
+ "name": "MatAnyone_encoder.mlpackage.zip",
712
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/matanyone/MatAnyone_encoder.mlpackage.zip",
713
  "archive": "zip",
714
+ "size_bytes": 17306121,
715
+ "sha256": "97ffd6bc4611f9a3351dc890fc00954ba48171e517e66a39f7a5f1f38110dfda",
716
  "compute_units": "all",
717
  "kind": "model"
718
  },
719
  {
720
+ "name": "MatAnyone_mask_encoder.mlpackage.zip",
721
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/matanyone/MatAnyone_mask_encoder.mlpackage.zip",
722
  "archive": "zip",
723
+ "size_bytes": 16819866,
724
+ "sha256": "ba67559188ffc64d8e46418c051c6a55815d4482def17519fa518daac7d5a911",
725
  "compute_units": "all",
726
  "kind": "model"
727
  },
728
  {
729
+ "name": "MatAnyone_read_first.mlpackage.zip",
730
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/matanyone/MatAnyone_read_first.mlpackage.zip",
731
  "archive": "zip",
732
+ "size_bytes": 21991849,
733
+ "sha256": "34daf7227dbcec7373a3fef175259fa7ec631ed8cb91d5595ca57ee9b22df7bb",
734
  "compute_units": "cpuOnly",
735
  "kind": "model"
736
  },
737
  {
738
+ "name": "MatAnyone_read.mlpackage.zip",
739
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/matanyone/MatAnyone_read.mlpackage.zip",
740
  "archive": "zip",
741
+ "size_bytes": 22135429,
742
+ "sha256": "052e52c0ffb7ff9ede448128950cd4c1c9a96589b6900c82b5104d99addb7fa5",
743
  "compute_units": "cpuOnly",
744
  "kind": "model"
745
  },
746
  {
747
+ "name": "MatAnyone_decoder.mlpackage.zip",
748
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/matanyone/MatAnyone_decoder.mlpackage.zip",
749
  "archive": "zip",
750
+ "size_bytes": 8807630,
751
+ "sha256": "67136aa67000e604838fe9aa7de151c514ef84f0b83f1da0f043cf70652d28eb",
752
  "compute_units": "all",
753
  "kind": "model"
754
  }
755
  ],
756
+ "requirements": {
757
+ "min_ios": "17.0",
758
+ "min_ram_mb": 800
759
+ },
760
+ "license": {
761
+ "name": "MIT",
762
+ "url": "https://github.com/pq-yang/MatAnyone"
763
+ },
764
+ "upstream": {
765
+ "name": "pq-yang/MatAnyone",
766
+ "url": "https://github.com/pq-yang/MatAnyone",
767
+ "year": 2025
768
+ }
769
  },
770
  {
771
  "id": "demucs",
 
778
  "config": {
779
  "sample_rate": 44100,
780
  "segment_length": 343980,
781
+ "output_stems": [
782
+ "drums",
783
+ "bass",
784
+ "vocals",
785
+ "other"
786
+ ]
787
  }
788
  },
789
  "files": [
790
  {
791
  "name": "HTDemucs_SourceSeparation_F32.mlpackage.zip",
792
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/demucs/HTDemucs_SourceSeparation_F32.mlpackage.zip",
793
  "archive": "zip",
794
+ "size_bytes": 79076395,
795
+ "sha256": "0fbb941e15a5b2fa425d14fe630ed4c14b6dee72780c1f5b2b05f58803bce5f7",
796
  "compute_units": "cpuOnly",
797
  "kind": "model"
798
  }
799
  ],
800
+ "requirements": {
801
+ "min_ios": "17.0",
802
+ "min_ram_mb": 1000
803
+ },
804
+ "license": {
805
+ "name": "MIT",
806
+ "url": "https://github.com/adefossez/demucs"
807
+ },
808
+ "upstream": {
809
+ "name": "adefossez/demucs",
810
+ "url": "https://github.com/adefossez/demucs",
811
+ "year": 2021
812
+ }
813
  },
814
  {
815
  "id": "kokoro",
 
823
  "mode": "tts",
824
  "sample_rate": 24000,
825
  "vocab_file": "kokoro_vocab.json",
826
+ "voices": [
827
+ "af_heart",
828
+ "af_bella",
829
+ "am_michael",
830
+ "bf_emma",
831
+ "bm_george"
832
+ ]
833
  }
834
  },
835
  "files": [
836
  {
837
  "name": "Kokoro_Predictor.mlpackage.zip",
838
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/kokoro/Kokoro_Predictor.mlpackage.zip",
839
  "archive": "zip",
840
+ "size_bytes": 72191470,
841
+ "sha256": "af1d55dc842980c32b5591a70f603941f11ab60a435bed0c13a107a8ef467bed",
842
  "compute_units": "cpuAndGPU",
843
  "kind": "model"
844
  },
845
  {
846
  "name": "Kokoro_Decoder_128.mlpackage.zip",
847
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/kokoro/Kokoro_Decoder_128.mlpackage.zip",
848
  "archive": "zip",
849
+ "size_bytes": 229120589,
850
+ "sha256": "cece0d072f5ba6aa3f729cf4c76b4de51823bcc65a26ab363c10441c3cd8b306",
851
  "compute_units": "all",
852
  "kind": "model"
853
  },
854
  {
855
  "name": "Kokoro_Decoder_256.mlpackage.zip",
856
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/kokoro/Kokoro_Decoder_256.mlpackage.zip",
857
  "archive": "zip",
858
+ "size_bytes": 229123438,
859
+ "sha256": "36d5e16d5c5ccb500fc96f1b07a1d5ac57b791f8e09e61b78319d76949003efe",
860
  "compute_units": "all",
861
  "kind": "model"
862
  },
863
  {
864
  "name": "Kokoro_Decoder_512.mlpackage.zip",
865
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/kokoro/Kokoro_Decoder_512.mlpackage.zip",
866
  "archive": "zip",
867
+ "size_bytes": 229128735,
868
+ "sha256": "0a44484c327e4fe8443b0bcf104d6964fe3f30d628c9e78aee3f31af7f2475dc",
869
  "compute_units": "all",
870
  "kind": "model"
871
  },
872
  {
873
  "name": "kokoro_vocab.json",
874
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/kokoro/kokoro_vocab.json",
875
+ "size_bytes": 1144,
876
+ "sha256": "70abefbe8a1c8865e43e0a43bbdc25b91a33e4aa053479d443ccf23e20a59e5d",
877
  "kind": "vocab"
878
  }
879
  ],
880
+ "requirements": {
881
+ "min_ios": "17.0",
882
+ "min_ram_mb": 1000
883
+ },
884
+ "license": {
885
+ "name": "Apache-2.0",
886
+ "url": "https://huggingface.co/hexgrad/Kokoro-82M"
887
+ },
888
+ "upstream": {
889
+ "name": "hexgrad/Kokoro-82M",
890
+ "url": "https://huggingface.co/hexgrad/Kokoro-82M",
891
+ "year": 2024
892
+ }
893
  },
894
  {
895
  "id": "stable_audio",
 
908
  "files": [
909
  {
910
  "name": "StableAudioT5Encoder.mlpackage.zip",
911
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/stableaudio/StableAudioT5Encoder.mlpackage.zip",
912
  "archive": "zip",
913
+ "size_bytes": 98538259,
914
+ "sha256": "319a8ba775d309240253ced68a03a3923d0aec9a79f608044f9403bdcfe4b741",
915
  "compute_units": "cpuOnly",
916
  "kind": "model"
917
  },
918
  {
919
  "name": "StableAudioNumberEmbedder.mlpackage.zip",
920
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/stableaudio/StableAudioNumberEmbedder.mlpackage.zip",
921
  "archive": "zip",
922
+ "size_bytes": 376018,
923
+ "sha256": "04bdc5de00a2cf1c4a18f80c94f0d74ecfab41f3ad99f2fb7a031d6ff5af75da",
924
  "compute_units": "cpuOnly",
925
  "kind": "model"
926
  },
927
  {
928
  "name": "StableAudioDiT.mlpackage.zip",
929
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/stableaudio/StableAudioDiT.mlpackage.zip",
930
  "archive": "zip",
931
+ "size_bytes": 1265748504,
932
+ "sha256": "b17da4fc4df857821d39dbdf7d3bfe7062a2272ab3e5df1284d545afb54047e4",
933
  "compute_units": "cpuOnly",
934
  "kind": "model"
935
  },
936
  {
937
  "name": "StableAudioVAEDecoder.mlpackage.zip",
938
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/stableaudio/StableAudioVAEDecoder.mlpackage.zip",
939
  "archive": "zip",
940
+ "size_bytes": 144960275,
941
+ "sha256": "7207544cca9799cc1d6803c5e81badd0bb4527b2d3a64d5cab5700a5f19a9374",
942
  "compute_units": "cpuAndGPU",
943
  "kind": "model"
944
  },
945
  {
946
  "name": "t5_vocab.json",
947
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/stableaudio/t5_vocab.json",
948
+ "size_bytes": 749757,
949
+ "sha256": "7c9ff3ac1b3dbcaa617ee659f2df68688cfd44f1a5eb3be3fa0a2f56c749d56a",
950
  "kind": "vocab"
951
  }
952
  ],
953
+ "requirements": {
954
+ "min_ios": "17.0",
955
+ "min_ram_mb": 1200
956
+ },
957
+ "license": {
958
+ "name": "custom",
959
+ "url": "https://huggingface.co/stabilityai/stable-audio-open-small"
960
+ },
961
+ "upstream": {
962
+ "name": "stabilityai/stable-audio-open-small",
963
+ "url": "https://huggingface.co/stabilityai/stable-audio-open-small",
964
+ "year": 2024
965
+ }
966
  },
967
  {
968
  "id": "basicpitch",
 
992
  "kind": "model"
993
  }
994
  ],
995
+ "requirements": {
996
+ "min_ios": "17.0",
997
+ "min_ram_mb": 200
998
+ },
999
+ "license": {
1000
+ "name": "Apache-2.0",
1001
+ "url": "https://github.com/spotify/basic-pitch"
1002
+ },
1003
+ "upstream": {
1004
+ "name": "spotify/basic-pitch",
1005
+ "url": "https://github.com/spotify/basic-pitch",
1006
+ "year": 2022
1007
+ }
1008
  },
1009
  {
1010
  "id": "diarization",
 
1016
  "template": "audio_in_out",
1017
  "config": {
1018
  "sample_rate": 16000,
1019
+ "output_stems": [
1020
+ "speaker_timeline"
1021
+ ]
1022
  }
1023
  },
1024
  "files": [
1025
  {
1026
+ "name": "SpeakerSegmentation.mlpackage.zip",
1027
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/diarization/SpeakerSegmentation.mlpackage.zip",
1028
  "archive": "zip",
1029
+ "size_bytes": 5327137,
1030
+ "sha256": "dcfa2b98900f2b99029abfb593644b70418186a6ec2e94c9a79c2b3d7a84378a",
1031
  "compute_units": "cpuAndGPU",
1032
  "kind": "model"
1033
  }
1034
  ],
1035
+ "requirements": {
1036
+ "min_ios": "17.0",
1037
+ "min_ram_mb": 200
1038
+ },
1039
+ "license": {
1040
+ "name": "MIT",
1041
+ "url": "https://github.com/pyannote/pyannote-audio"
1042
+ },
1043
+ "upstream": {
1044
+ "name": "pyannote/pyannote-audio",
1045
+ "url": "https://github.com/pyannote/pyannote-audio",
1046
+ "year": 2021
1047
+ }
1048
  },
1049
  {
1050
  "id": "openvoice",
 
1056
  "template": "audio_in_out",
1057
  "config": {
1058
  "sample_rate": 22050,
1059
+ "output_stems": [
1060
+ "converted"
1061
+ ]
1062
  }
1063
  },
1064
  "files": [
1065
  {
1066
  "name": "OpenVoice_SpeakerEncoder.mlpackage.zip",
1067
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/openvoice/OpenVoice_SpeakerEncoder.mlpackage.zip",
1068
  "archive": "zip",
1069
+ "size_bytes": 1519880,
1070
+ "sha256": "c3f2a96aaf5ecb5c5afc62b3d3dfbd47dc7ae64bc3edb7aa68befb54aef74459",
1071
  "compute_units": "cpuAndGPU",
1072
  "kind": "model"
1073
  },
1074
  {
1075
  "name": "OpenVoice_VoiceConverter.mlpackage.zip",
1076
+ "url": "https://huggingface.co/mlboydaisuke/coreml-zoo/resolve/main/openvoice/OpenVoice_VoiceConverter.mlpackage.zip",
1077
  "archive": "zip",
1078
+ "size_bytes": 59799630,
1079
+ "sha256": "ef3ce8a2d1564aefa13830d7d0ca43f85e0aa62d5f59622c8bc456c307ab5e05",
1080
  "compute_units": "cpuAndGPU",
1081
  "kind": "model"
1082
  }
1083
  ],
1084
+ "requirements": {
1085
+ "min_ios": "17.0",
1086
+ "min_ram_mb": 500
1087
+ },
1088
+ "license": {
1089
+ "name": "MIT",
1090
+ "url": "https://github.com/myshell-ai/OpenVoice"
1091
+ },
1092
+ "upstream": {
1093
+ "name": "myshell-ai/OpenVoice",
1094
+ "url": "https://github.com/myshell-ai/OpenVoice",
1095
+ "year": 2023
1096
+ }
1097
  },
1098
  {
1099
  "id": "realesrgan",
 
1103
  "description_md": "Real-world blind super-resolution. 4× upscale from any input. Handles noise, blur, and JPEG artifacts. 512×512 input → 2048×2048 output.",
1104
  "demo": {
1105
  "template": "image_in_out",
1106
+ "config": {
1107
+ "input_size": 512,
1108
+ "output_type": "image"
1109
+ }
1110
  },
1111
  "files": [
1112
  {
 
1119
  "kind": "model"
1120
  }
1121
  ],
1122
+ "requirements": {
1123
+ "min_ios": "17.0",
1124
+ "min_ram_mb": 500
1125
+ },
1126
+ "license": {
1127
+ "name": "BSD-3-Clause",
1128
+ "url": "https://github.com/xinntao/Real-ESRGAN/blob/master/LICENSE"
1129
+ },
1130
+ "upstream": {
1131
+ "name": "xinntao/Real-ESRGAN",
1132
+ "url": "https://github.com/xinntao/Real-ESRGAN",
1133
+ "year": 2021
1134
+ }
1135
  },
1136
  {
1137
  "id": "gfpgan",
 
1141
  "description_md": "Blind face restoration with generative facial prior. Restores degraded face photos to high quality. 512×512 input/output.",
1142
  "demo": {
1143
  "template": "image_in_out",
1144
+ "config": {
1145
+ "input_size": 512,
1146
+ "output_type": "image"
1147
+ }
1148
  },
1149
  "files": [
1150
  {
 
1157
  "kind": "model"
1158
  }
1159
  ],
1160
+ "requirements": {
1161
+ "min_ios": "17.0",
1162
+ "min_ram_mb": 600
1163
+ },
1164
+ "license": {
1165
+ "name": "Apache-2.0",
1166
+ "url": "https://github.com/TencentARC/GFPGAN/blob/master/LICENSE"
1167
+ },
1168
+ "upstream": {
1169
+ "name": "TencentARC/GFPGAN",
1170
+ "url": "https://github.com/TencentARC/GFPGAN",
1171
+ "year": 2021
1172
+ }
1173
  },
1174
  {
1175
  "id": "rfdetr_n",
 
1198
  "kind": "model"
1199
  }
1200
  ],
1201
+ "requirements": {
1202
+ "min_ios": "17.0",
1203
+ "min_ram_mb": 400
1204
+ },
1205
+ "license": {
1206
+ "name": "Apache-2.0",
1207
+ "url": "https://github.com/roboflow/rf-detr/blob/main/LICENSE"
1208
+ },
1209
+ "upstream": {
1210
+ "name": "roboflow/rf-detr",
1211
+ "url": "https://github.com/roboflow/rf-detr",
1212
+ "year": 2025
1213
+ }
1214
  },
1215
  {
1216
  "id": "face_parsing",
 
1220
  "description_md": "Semantic face parsing into 19 regions: skin, nose, eyes, eyebrows, ears, mouth, lip, hair, hat, eyeglass, earring, necklace, neck, cloth, background. 512×512 input.",
1221
  "demo": {
1222
  "template": "image_in_out",
1223
+ "config": {
1224
+ "input_size": 512,
1225
+ "output_type": "segmap",
1226
+ "num_classes": 19
1227
+ }
1228
  },
1229
  "files": [
1230
  {
 
1237
  "kind": "model"
1238
  }
1239
  ],
1240
+ "requirements": {
1241
+ "min_ios": "17.0",
1242
+ "min_ram_mb": 300
1243
+ },
1244
+ "license": {
1245
+ "name": "MIT",
1246
+ "url": "https://github.com/zllrunning/face-parsing.PyTorch/blob/master/LICENSE"
1247
+ },
1248
+ "upstream": {
1249
+ "name": "zllrunning/face-parsing.PyTorch",
1250
+ "url": "https://github.com/zllrunning/face-parsing.PyTorch",
1251
+ "year": 2019
1252
+ }
1253
  },
1254
  {
1255
  "id": "mobilesam",
 
1276
  "kind": "model"
1277
  }
1278
  ],
1279
+ "requirements": {
1280
+ "min_ios": "17.0",
1281
+ "min_ram_mb": 300
1282
+ },
1283
+ "license": {
1284
+ "name": "Apache-2.0",
1285
+ "url": "https://github.com/ChaoningZhang/MobileSAM/blob/master/LICENSE"
1286
+ },
1287
+ "upstream": {
1288
+ "name": "ChaoningZhang/MobileSAM",
1289
+ "url": "https://github.com/ChaoningZhang/MobileSAM",
1290
+ "year": 2023
1291
+ }
1292
  },
1293
  {
1294
  "id": "lama",
 
1298
  "description_md": "Resolution-robust large mask inpainting. Draw over unwanted objects to remove them. Fast Fourier convolutions for global context. 800×800 input.",
1299
  "demo": {
1300
  "template": "inpainting",
1301
+ "config": {
1302
+ "input_size": 800
1303
+ }
1304
  },
1305
  "files": [
1306
  {
 
1313
  "kind": "model"
1314
  }
1315
  ],
1316
+ "requirements": {
1317
+ "min_ios": "17.0",
1318
+ "min_ram_mb": 600
1319
+ },
1320
+ "license": {
1321
+ "name": "Apache-2.0",
1322
+ "url": "https://github.com/advimman/lama/blob/main/LICENSE"
1323
+ },
1324
+ "upstream": {
1325
+ "name": "advimman/lama",
1326
+ "url": "https://github.com/advimman/lama",
1327
+ "year": 2022
1328
+ }
1329
  }
1330
  ]
1331
  }