Coffmeteor commited on
Commit
3f331d7
·
verified ·
1 Parent(s): da88848

Mirror published Civitai release: Storyflow Character Card

Browse files
storyflow/character-card-v0.4/README.md ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Storyflow Character Card
2
+
3
+ Published by **Coffmeteor** for **Krea 2**.
4
+
5
+ [Original Civitai release](https://civitai.red/models/2906958?modelVersionId=3314189) · Version: v0.4 — Three-View Character Card
6
+
7
+ The files below preserve the exact published Civitai bytes. SHA-256 checksums are included.
8
+
9
+ ## Downloads
10
+
11
+ | File | Type |
12
+ |---|---|
13
+ | [Storyflow-v04-spatial-node-encoder-fix.zip](Storyflow-v04-spatial-node-encoder-fix.zip) | Other |
14
+ | [Storyflow-v04-support-BF16-default.zip](Storyflow-v04-support-BF16-default.zip) | Other |
15
+ | [Storyflow-v04-form-BF16-default.json](Storyflow-v04-form-BF16-default.json) | Config |
16
+ | [Storyflow-v04-standard-BF16-default.json](Storyflow-v04-standard-BF16-default.json) | Config |
17
+ | [candidate320-single-bf16.safetensors](candidate320-single-bf16.safetensors) | Model |
18
+ | [candidate320-single.safetensors](candidate320-single.safetensors) | Model |
19
+
20
+ ## Usage
21
+
22
+ The Civitai trigger field is empty; follow the published usage notes below.
23
+
24
+ Verified default adapter strength: 1.
25
+
26
+ This is the current v0.4 spatial/reference-conditioned release. Use its included support package and node, retain the BF16 workflow defaults, and upload your own face reference. Both BF16 and FP32 adapter files are provided. Supported output sizes: 1024×576, 1280×720, and 1536×864. Earlier v0.1/v0.2 adapters are superseded.
27
+
28
+ ## Provenance and terms
29
+
30
+ The original public descriptions, exact trigger field, permission flags and file metadata are retained in [source.json](source.json). Original model and component terms remain applicable; this mirror does not grant a different license.
31
+
32
+ [SHA-256 checksums](SHA256SUMS)
storyflow/character-card-v0.4/SHA256SUMS ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ 7e4ad3f584886e871015014a4156d01633578fe12df974fa170ac15407171604 Storyflow-v04-spatial-node-encoder-fix.zip
2
+ 0dc8ddbaf29661e8c84c7c81019e11cfa3a0ddc390050f299ef03cb294c5869f Storyflow-v04-support-BF16-default.zip
3
+ 89e3edfb8c75b2fbb0a1bfd7adb28549baf60c77acddaedebd06564e548d0d65 Storyflow-v04-form-BF16-default.json
4
+ 7e31c93e8b3256633861b3e9e4fa934ca4e8b7a2b168ca76f90e45dd549b0134 Storyflow-v04-standard-BF16-default.json
5
+ 17d9ae50d2714b7c0d04da9a4f4f5755ac775dc28deb628a4fa51f1130d7b2ad candidate320-single-bf16.safetensors
6
+ 2bf21aa6ddb781a5d12710e9a4930671d1d98f8f84c546d749c76f5865898647 candidate320-single.safetensors
storyflow/character-card-v0.4/Storyflow-v04-form-BF16-default.json ADDED
@@ -0,0 +1,1515 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "id": "3579f887-385c-58eb-b676-e4c392bf0fd7",
3
+ "revision": 0,
4
+ "last_node_id": 920,
5
+ "last_link_id": 35,
6
+ "nodes": [
7
+ {
8
+ "id": 55,
9
+ "type": "UNETLoader",
10
+ "pos": [
11
+ 20,
12
+ 1180
13
+ ],
14
+ "size": [
15
+ 360,
16
+ 130
17
+ ],
18
+ "flags": {
19
+ "collapsed": true
20
+ },
21
+ "order": 0,
22
+ "mode": 0,
23
+ "inputs": [],
24
+ "outputs": [
25
+ {
26
+ "name": "MODEL",
27
+ "type": "MODEL",
28
+ "links": [
29
+ 10
30
+ ],
31
+ "slot_index": 0
32
+ }
33
+ ],
34
+ "properties": {
35
+ "Node name for S&R": "UNETLoader"
36
+ },
37
+ "widgets_values": [
38
+ "krea2_turbo_bf16.safetensors",
39
+ "default"
40
+ ],
41
+ "title": "Krea2 模型"
42
+ },
43
+ {
44
+ "id": 56,
45
+ "type": "CharacterCardGroundedTextEncoder",
46
+ "pos": [
47
+ 360,
48
+ 1180
49
+ ],
50
+ "size": [
51
+ 470,
52
+ 130
53
+ ],
54
+ "flags": {
55
+ "collapsed": true
56
+ },
57
+ "order": 1,
58
+ "mode": 0,
59
+ "inputs": [],
60
+ "outputs": [
61
+ {
62
+ "name": "clip",
63
+ "type": "CLIP",
64
+ "links": [
65
+ 5,
66
+ 8,
67
+ 26
68
+ ],
69
+ "slot_index": 0
70
+ }
71
+ ],
72
+ "properties": {
73
+ "Node name for S&R": "CharacterCardGroundedTextEncoder"
74
+ },
75
+ "widgets_values": [
76
+ "character-card/qwen3vl_training_bf16.safetensors"
77
+ ],
78
+ "title": "文本编码器 / Text encoder · BF16 default, FP8 optional"
79
+ },
80
+ {
81
+ "id": 57,
82
+ "type": "VAELoader",
83
+ "pos": [
84
+ 700,
85
+ 1180
86
+ ],
87
+ "size": [
88
+ 390,
89
+ 130
90
+ ],
91
+ "flags": {
92
+ "collapsed": true
93
+ },
94
+ "order": 2,
95
+ "mode": 0,
96
+ "inputs": [],
97
+ "outputs": [
98
+ {
99
+ "name": "VAE",
100
+ "type": "VAE",
101
+ "links": [
102
+ 2,
103
+ 12,
104
+ 23
105
+ ],
106
+ "slot_index": 0
107
+ }
108
+ ],
109
+ "properties": {
110
+ "Node name for S&R": "VAELoader"
111
+ },
112
+ "widgets_values": [
113
+ "qwen/qwen_image_vae.safetensors"
114
+ ],
115
+ "title": "VAE"
116
+ },
117
+ {
118
+ "id": 72,
119
+ "type": "LoadImage",
120
+ "pos": [
121
+ 20,
122
+ 20
123
+ ],
124
+ "size": [
125
+ 330,
126
+ 430
127
+ ],
128
+ "flags": {},
129
+ "order": 3,
130
+ "mode": 0,
131
+ "inputs": [
132
+ {
133
+ "name": "image",
134
+ "type": "COMBO",
135
+ "link": null,
136
+ "widget": {
137
+ "name": "image"
138
+ },
139
+ "label": "参考图 / Reference"
140
+ }
141
+ ],
142
+ "outputs": [
143
+ {
144
+ "name": "IMAGE",
145
+ "type": "IMAGE",
146
+ "links": [
147
+ 1,
148
+ 6,
149
+ 9,
150
+ 13
151
+ ],
152
+ "slot_index": 0
153
+ },
154
+ {
155
+ "name": "MASK",
156
+ "type": "MASK",
157
+ "links": [],
158
+ "slot_index": 1
159
+ }
160
+ ],
161
+ "properties": {
162
+ "Node name for S&R": "LoadImage"
163
+ },
164
+ "widgets_values": [
165
+ "storyflow-v03/df34f46583e7d1ebf445e93db03c15a533155c096075a043cf351a42d8fcb181.png",
166
+ "image"
167
+ ],
168
+ "title": "① 参考图 / Reference · Upload photo"
169
+ },
170
+ {
171
+ "id": 73,
172
+ "type": "VAEEncode",
173
+ "pos": [
174
+ 1380,
175
+ 1180
176
+ ],
177
+ "size": [
178
+ 340,
179
+ 120
180
+ ],
181
+ "flags": {
182
+ "collapsed": true
183
+ },
184
+ "order": 4,
185
+ "mode": 0,
186
+ "inputs": [
187
+ {
188
+ "name": "pixels",
189
+ "type": "IMAGE",
190
+ "link": 1
191
+ },
192
+ {
193
+ "name": "vae",
194
+ "type": "VAE",
195
+ "link": 2
196
+ }
197
+ ],
198
+ "outputs": [
199
+ {
200
+ "name": "LATENT",
201
+ "type": "LATENT",
202
+ "links": [
203
+ 11
204
+ ],
205
+ "slot_index": 0
206
+ }
207
+ ],
208
+ "properties": {
209
+ "Node name for S&R": "VAEEncode"
210
+ },
211
+ "widgets_values": [],
212
+ "title": "参考 VAE 编码"
213
+ },
214
+ {
215
+ "id": 82,
216
+ "type": "EmptySD3LatentImage",
217
+ "pos": [
218
+ 20,
219
+ 1280
220
+ ],
221
+ "size": [
222
+ 370,
223
+ 150
224
+ ],
225
+ "flags": {
226
+ "collapsed": true
227
+ },
228
+ "order": 9,
229
+ "mode": 0,
230
+ "inputs": [
231
+ {
232
+ "name": "width",
233
+ "type": "INT",
234
+ "link": 3,
235
+ "widget": {
236
+ "name": "width"
237
+ }
238
+ },
239
+ {
240
+ "name": "height",
241
+ "type": "INT",
242
+ "link": 4,
243
+ "widget": {
244
+ "name": "height"
245
+ }
246
+ }
247
+ ],
248
+ "outputs": [
249
+ {
250
+ "name": "LATENT",
251
+ "type": "LATENT",
252
+ "links": [
253
+ 14,
254
+ 18,
255
+ 28
256
+ ],
257
+ "slot_index": 0
258
+ }
259
+ ],
260
+ "properties": {
261
+ "Node name for S&R": "EmptySD3LatentImage"
262
+ },
263
+ "widgets_values": [
264
+ 1024,
265
+ 576,
266
+ 1
267
+ ],
268
+ "title": "③ 画布 · width / height"
269
+ },
270
+ {
271
+ "id": 84,
272
+ "type": "Krea2EditGroundedEncode",
273
+ "pos": [
274
+ 360,
275
+ 1280
276
+ ],
277
+ "size": [
278
+ 460,
279
+ 260
280
+ ],
281
+ "flags": {
282
+ "collapsed": true
283
+ },
284
+ "order": 10,
285
+ "mode": 0,
286
+ "inputs": [
287
+ {
288
+ "name": "clip",
289
+ "type": "CLIP",
290
+ "link": 5
291
+ },
292
+ {
293
+ "name": "image",
294
+ "type": "IMAGE",
295
+ "link": 6
296
+ },
297
+ {
298
+ "name": "prompt",
299
+ "type": "STRING",
300
+ "link": 7,
301
+ "widget": {
302
+ "name": "prompt"
303
+ }
304
+ }
305
+ ],
306
+ "outputs": [
307
+ {
308
+ "name": "CONDITIONING",
309
+ "type": "CONDITIONING",
310
+ "links": [
311
+ 27
312
+ ],
313
+ "slot_index": 0
314
+ }
315
+ ],
316
+ "properties": {
317
+ "Node name for S&R": "Krea2EditGroundedEncode"
318
+ },
319
+ "widgets_values": [
320
+ "",
321
+ 768,
322
+ ""
323
+ ],
324
+ "title": "全局描述 + 同一参考图"
325
+ },
326
+ {
327
+ "id": 85,
328
+ "type": "Krea2EditGroundedEncode",
329
+ "pos": [
330
+ 700,
331
+ 1280
332
+ ],
333
+ "size": [
334
+ 460,
335
+ 260
336
+ ],
337
+ "flags": {
338
+ "collapsed": true
339
+ },
340
+ "order": 11,
341
+ "mode": 0,
342
+ "inputs": [
343
+ {
344
+ "name": "clip",
345
+ "type": "CLIP",
346
+ "link": 8
347
+ },
348
+ {
349
+ "name": "image",
350
+ "type": "IMAGE",
351
+ "link": 9
352
+ }
353
+ ],
354
+ "outputs": [
355
+ {
356
+ "name": "CONDITIONING",
357
+ "type": "CONDITIONING",
358
+ "links": [
359
+ 17
360
+ ],
361
+ "slot_index": 0
362
+ }
363
+ ],
364
+ "title": "空负向 · CFG 1",
365
+ "properties": {
366
+ "Node name for S&R": "Krea2EditGroundedEncode"
367
+ },
368
+ "widgets_values": [
369
+ "",
370
+ 768,
371
+ ""
372
+ ]
373
+ },
374
+ {
375
+ "id": 79,
376
+ "type": "Krea2EditModelPatch",
377
+ "pos": [
378
+ 1040,
379
+ 1280
380
+ ],
381
+ "size": [
382
+ 390,
383
+ 260
384
+ ],
385
+ "flags": {
386
+ "collapsed": true
387
+ },
388
+ "order": 12,
389
+ "mode": 0,
390
+ "inputs": [
391
+ {
392
+ "name": "model",
393
+ "type": "MODEL",
394
+ "link": 10
395
+ },
396
+ {
397
+ "name": "source_latent",
398
+ "type": "LATENT",
399
+ "link": 11
400
+ },
401
+ {
402
+ "name": "vae",
403
+ "type": "VAE",
404
+ "link": 12
405
+ },
406
+ {
407
+ "name": "source_image",
408
+ "type": "IMAGE",
409
+ "link": 13
410
+ },
411
+ {
412
+ "name": "target_latent",
413
+ "type": "LATENT",
414
+ "link": 14
415
+ }
416
+ ],
417
+ "outputs": [
418
+ {
419
+ "name": "MODEL",
420
+ "type": "MODEL",
421
+ "links": [
422
+ 25
423
+ ],
424
+ "slot_index": 0
425
+ }
426
+ ],
427
+ "properties": {
428
+ "Node name for S&R": "Krea2EditModelPatch"
429
+ },
430
+ "widgets_values": [
431
+ 1.0,
432
+ 1.0,
433
+ "fit"
434
+ ],
435
+ "title": "参考条件 · fit / strength 1"
436
+ },
437
+ {
438
+ "id": 53,
439
+ "type": "SamplerCustom",
440
+ "pos": [
441
+ 1380,
442
+ 1280
443
+ ],
444
+ "size": [
445
+ 370,
446
+ 330
447
+ ],
448
+ "flags": {
449
+ "collapsed": true
450
+ },
451
+ "order": 16,
452
+ "mode": 0,
453
+ "inputs": [
454
+ {
455
+ "name": "model",
456
+ "type": "MODEL",
457
+ "link": 15
458
+ },
459
+ {
460
+ "name": "positive",
461
+ "type": "CONDITIONING",
462
+ "link": 16
463
+ },
464
+ {
465
+ "name": "negative",
466
+ "type": "CONDITIONING",
467
+ "link": 17
468
+ },
469
+ {
470
+ "name": "latent_image",
471
+ "type": "LATENT",
472
+ "link": 18
473
+ },
474
+ {
475
+ "name": "sampler",
476
+ "type": "SAMPLER",
477
+ "link": 19
478
+ },
479
+ {
480
+ "name": "sigmas",
481
+ "type": "SIGMAS",
482
+ "link": 20
483
+ },
484
+ {
485
+ "name": "noise_seed",
486
+ "type": "INT",
487
+ "link": 21,
488
+ "widget": {
489
+ "name": "noise_seed"
490
+ }
491
+ }
492
+ ],
493
+ "outputs": [
494
+ {
495
+ "name": "output",
496
+ "type": "LATENT",
497
+ "links": [
498
+ 30
499
+ ],
500
+ "slot_index": 0
501
+ },
502
+ {
503
+ "name": "denoised_output",
504
+ "type": "LATENT",
505
+ "links": [],
506
+ "slot_index": 1
507
+ }
508
+ ],
509
+ "properties": {
510
+ "Node name for S&R": "SamplerCustom"
511
+ },
512
+ "widgets_values": [
513
+ true,
514
+ 20260910,
515
+ "fixed",
516
+ 1.0
517
+ ],
518
+ "title": "④ 种子 · fixed / CFG 1"
519
+ },
520
+ {
521
+ "id": 54,
522
+ "type": "VAEDecode",
523
+ "pos": [
524
+ 20,
525
+ 1380
526
+ ],
527
+ "size": [
528
+ 330,
529
+ 150
530
+ ],
531
+ "flags": {
532
+ "collapsed": true
533
+ },
534
+ "order": 18,
535
+ "mode": 0,
536
+ "inputs": [
537
+ {
538
+ "name": "samples",
539
+ "type": "LATENT",
540
+ "link": 22
541
+ },
542
+ {
543
+ "name": "vae",
544
+ "type": "VAE",
545
+ "link": 23
546
+ }
547
+ ],
548
+ "outputs": [
549
+ {
550
+ "name": "IMAGE",
551
+ "type": "IMAGE",
552
+ "links": [
553
+ 24
554
+ ],
555
+ "slot_index": 0
556
+ }
557
+ ],
558
+ "properties": {
559
+ "Node name for S&R": "VAEDecode"
560
+ },
561
+ "widgets_values": [],
562
+ "title": "原画布解码"
563
+ },
564
+ {
565
+ "id": 29,
566
+ "type": "SaveImage",
567
+ "pos": [
568
+ 950,
569
+ 20
570
+ ],
571
+ "size": [
572
+ 670,
573
+ 790
574
+ ],
575
+ "flags": {},
576
+ "order": 19,
577
+ "mode": 0,
578
+ "inputs": [
579
+ {
580
+ "name": "images",
581
+ "type": "IMAGE",
582
+ "link": 24
583
+ }
584
+ ],
585
+ "outputs": [
586
+ {
587
+ "name": "images",
588
+ "type": "IMAGE",
589
+ "links": [],
590
+ "slot_index": 0
591
+ }
592
+ ],
593
+ "properties": {
594
+ "Node name for S&R": "SaveImage"
595
+ },
596
+ "widgets_values": [
597
+ "storyflow-v04/candidate_native_card"
598
+ ],
599
+ "title": "⑥ 人物卡 / Character card · Preview & Save"
600
+ },
601
+ {
602
+ "id": 120,
603
+ "pos": [
604
+ 700,
605
+ 1380
606
+ ],
607
+ "size": [
608
+ 360,
609
+ 120
610
+ ],
611
+ "flags": {
612
+ "collapsed": true
613
+ },
614
+ "properties": {
615
+ "Node name for S&R": "KSamplerSelect"
616
+ },
617
+ "title": "Euler",
618
+ "type": "KSamplerSelect",
619
+ "order": 14,
620
+ "mode": 0,
621
+ "inputs": [],
622
+ "outputs": [
623
+ {
624
+ "name": "SAMPLER",
625
+ "type": "SAMPLER",
626
+ "links": [
627
+ 19
628
+ ],
629
+ "slot_index": 0
630
+ }
631
+ ],
632
+ "widgets_values": [
633
+ "euler"
634
+ ]
635
+ },
636
+ {
637
+ "id": 121,
638
+ "pos": [
639
+ 1040,
640
+ 1380
641
+ ],
642
+ "size": [
643
+ 470,
644
+ 220
645
+ ],
646
+ "flags": {
647
+ "collapsed": true
648
+ },
649
+ "properties": {
650
+ "Node name for S&R": "ManualSigmas"
651
+ },
652
+ "title": "固定 8 步 sigmas",
653
+ "type": "ManualSigmas",
654
+ "order": 15,
655
+ "mode": 0,
656
+ "inputs": [],
657
+ "outputs": [
658
+ {
659
+ "name": "SIGMAS",
660
+ "type": "SIGMAS",
661
+ "links": [
662
+ 20
663
+ ],
664
+ "slot_index": 0
665
+ }
666
+ ],
667
+ "widgets_values": [
668
+ "1.0,0.9248116147173576,0.8367657420590665,0.7322560154608024,0.6061882408399552,0.4511248963112157,0.25576263215491235,0.0020454536906595946,0.0"
669
+ ]
670
+ },
671
+ {
672
+ "id": 180,
673
+ "type": "CharacterCardGroundedSpatial",
674
+ "flags": {
675
+ "collapsed": true
676
+ },
677
+ "mode": 0,
678
+ "order": 13,
679
+ "inputs": [
680
+ {
681
+ "name": "model",
682
+ "type": "MODEL",
683
+ "link": 25
684
+ },
685
+ {
686
+ "name": "clip",
687
+ "type": "CLIP",
688
+ "link": 26
689
+ },
690
+ {
691
+ "name": "global_grounded",
692
+ "type": "CONDITIONING",
693
+ "link": 27
694
+ },
695
+ {
696
+ "name": "target_latent",
697
+ "type": "LATENT",
698
+ "link": 28
699
+ },
700
+ {
701
+ "name": "regions_json",
702
+ "type": "STRING",
703
+ "link": 29,
704
+ "widget": {
705
+ "name": "regions_json"
706
+ }
707
+ }
708
+ ],
709
+ "outputs": [
710
+ {
711
+ "name": "model",
712
+ "type": "MODEL",
713
+ "links": [
714
+ 15
715
+ ],
716
+ "slot_index": 0
717
+ },
718
+ {
719
+ "name": "conditioning",
720
+ "type": "CONDITIONING",
721
+ "links": [
722
+ 16
723
+ ],
724
+ "slot_index": 1
725
+ },
726
+ {
727
+ "name": "state",
728
+ "type": "GROUNDED_SPATIAL_STATE",
729
+ "links": [
730
+ 31
731
+ ],
732
+ "slot_index": 2
733
+ }
734
+ ],
735
+ "properties": {
736
+ "Node name for S&R": "CharacterCardGroundedSpatial"
737
+ },
738
+ "widgets_values": [
739
+ "character-card-v04/candidate320-single.safetensors",
740
+ ""
741
+ ],
742
+ "pos": [
743
+ 1380,
744
+ 1380
745
+ ],
746
+ "size": [
747
+ 850,
748
+ 360
749
+ ],
750
+ "title": "候选空间权重 · 选择 task + spatial 单文件"
751
+ },
752
+ {
753
+ "id": 181,
754
+ "type": "CharacterCardGroundedSpatialReport",
755
+ "flags": {
756
+ "collapsed": true
757
+ },
758
+ "mode": 0,
759
+ "order": 17,
760
+ "inputs": [
761
+ {
762
+ "name": "samples",
763
+ "type": "LATENT",
764
+ "link": 30
765
+ },
766
+ {
767
+ "name": "state",
768
+ "type": "GROUNDED_SPATIAL_STATE",
769
+ "link": 31
770
+ }
771
+ ],
772
+ "outputs": [
773
+ {
774
+ "name": "samples",
775
+ "type": "LATENT",
776
+ "links": [
777
+ 22
778
+ ],
779
+ "slot_index": 0
780
+ },
781
+ {
782
+ "name": "report",
783
+ "type": "STRING",
784
+ "links": [],
785
+ "slot_index": 1
786
+ }
787
+ ],
788
+ "properties": {
789
+ "Node name for S&R": "CharacterCardGroundedSpatialReport"
790
+ },
791
+ "widgets_values": [],
792
+ "pos": [
793
+ 20,
794
+ 1480
795
+ ],
796
+ "size": [
797
+ 330,
798
+ 180
799
+ ],
800
+ "title": "空间执行报告"
801
+ },
802
+ {
803
+ "id": 190,
804
+ "type": "CharacterCardFixedAppearanceInputs",
805
+ "flags": {},
806
+ "mode": 0,
807
+ "order": 8,
808
+ "inputs": [
809
+ {
810
+ "name": "outfit",
811
+ "type": "STRING",
812
+ "link": 32,
813
+ "widget": {
814
+ "name": "outfit"
815
+ }
816
+ },
817
+ {
818
+ "name": "body",
819
+ "type": "STRING",
820
+ "link": 33,
821
+ "widget": {
822
+ "name": "body"
823
+ }
824
+ },
825
+ {
826
+ "name": "styling",
827
+ "type": "STRING",
828
+ "link": 34,
829
+ "widget": {
830
+ "name": "styling"
831
+ }
832
+ },
833
+ {
834
+ "name": "resolution",
835
+ "type": "COMBO",
836
+ "link": null,
837
+ "widget": {
838
+ "name": "resolution"
839
+ },
840
+ "label": "输出尺寸 / Resolution"
841
+ },
842
+ {
843
+ "name": "seed",
844
+ "type": "INT",
845
+ "link": null,
846
+ "widget": {
847
+ "name": "seed"
848
+ },
849
+ "label": "种子 / Seed"
850
+ }
851
+ ],
852
+ "outputs": [
853
+ {
854
+ "name": "global_prompt",
855
+ "type": "STRING",
856
+ "links": [
857
+ 7
858
+ ],
859
+ "slot_index": 0
860
+ },
861
+ {
862
+ "name": "regions_json",
863
+ "type": "STRING",
864
+ "links": [
865
+ 29,
866
+ 35
867
+ ],
868
+ "slot_index": 1
869
+ },
870
+ {
871
+ "name": "width",
872
+ "type": "INT",
873
+ "links": [
874
+ 3
875
+ ],
876
+ "slot_index": 2
877
+ },
878
+ {
879
+ "name": "height",
880
+ "type": "INT",
881
+ "links": [
882
+ 4
883
+ ],
884
+ "slot_index": 3
885
+ },
886
+ {
887
+ "name": "seed",
888
+ "type": "INT",
889
+ "links": [
890
+ 21
891
+ ],
892
+ "slot_index": 4
893
+ }
894
+ ],
895
+ "properties": {
896
+ "Node name for S&R": "CharacterCardFixedAppearanceInputs"
897
+ },
898
+ "widgets_values": [
899
+ "",
900
+ "",
901
+ "",
902
+ "1024x576",
903
+ 20260910
904
+ ],
905
+ "pos": [
906
+ 20,
907
+ 520
908
+ ],
909
+ "size": [
910
+ 330,
911
+ 180
912
+ ],
913
+ "title": "⑤ 尺寸与种子 / Size & Seed"
914
+ },
915
+ {
916
+ "id": 192,
917
+ "type": "PrimitiveStringMultiline",
918
+ "flags": {},
919
+ "mode": 0,
920
+ "order": 5,
921
+ "inputs": [
922
+ {
923
+ "name": "value",
924
+ "type": "STRING",
925
+ "link": null,
926
+ "widget": {
927
+ "name": "value"
928
+ },
929
+ "label": "服装 / Outfit"
930
+ }
931
+ ],
932
+ "outputs": [
933
+ {
934
+ "name": "STRING",
935
+ "type": "STRING",
936
+ "links": [
937
+ 32
938
+ ],
939
+ "slot_index": 0
940
+ }
941
+ ],
942
+ "properties": {
943
+ "Node name for S&R": "PrimitiveStringMultiline"
944
+ },
945
+ "widgets_values": [
946
+ "An ochre tan cropped leather flight jacket with an ivory shearling collar, brass front zipper, ribbed cuffs and two angled flap pockets; an ivory ribbed turtleneck underneath; dark brown straight cargo trousers with modest thigh pockets and brown lace-up leather ankle boots. The back of the jacket has a plain horizontal shoulder yoke and a neat waist seam, no lettering."
947
+ ],
948
+ "pos": [
949
+ 400,
950
+ 20
951
+ ],
952
+ "size": [
953
+ 510,
954
+ 210
955
+ ],
956
+ "title": "② 服装 / Outfit · Colors, layers & shoes"
957
+ },
958
+ {
959
+ "id": 193,
960
+ "type": "PrimitiveStringMultiline",
961
+ "flags": {},
962
+ "mode": 0,
963
+ "order": 6,
964
+ "inputs": [
965
+ {
966
+ "name": "value",
967
+ "type": "STRING",
968
+ "link": null,
969
+ "widget": {
970
+ "name": "value"
971
+ },
972
+ "label": "身材 / Body"
973
+ }
974
+ ],
975
+ "outputs": [
976
+ {
977
+ "name": "STRING",
978
+ "type": "STRING",
979
+ "links": [
980
+ 33
981
+ ],
982
+ "slot_index": 0
983
+ }
984
+ ],
985
+ "properties": {
986
+ "Node name for S&R": "PrimitiveStringMultiline"
987
+ },
988
+ "widgets_values": [
989
+ "The person has a compact athletic build, sturdy shoulders, fit torso and naturally strong arms and legs, balanced realistic proportions."
990
+ ],
991
+ "pos": [
992
+ 400,
993
+ 290
994
+ ],
995
+ "size": [
996
+ 510,
997
+ 210
998
+ ],
999
+ "title": "③ 身材 / Body · Free-form description"
1000
+ },
1001
+ {
1002
+ "id": 194,
1003
+ "type": "PrimitiveStringMultiline",
1004
+ "flags": {},
1005
+ "mode": 0,
1006
+ "order": 7,
1007
+ "inputs": [
1008
+ {
1009
+ "name": "value",
1010
+ "type": "STRING",
1011
+ "link": null,
1012
+ "widget": {
1013
+ "name": "value"
1014
+ },
1015
+ "label": "风格 / Styling"
1016
+ }
1017
+ ],
1018
+ "outputs": [
1019
+ {
1020
+ "name": "STRING",
1021
+ "type": "STRING",
1022
+ "links": [
1023
+ 34
1024
+ ],
1025
+ "slot_index": 0
1026
+ }
1027
+ ],
1028
+ "properties": {
1029
+ "Node name for S&R": "PrimitiveStringMultiline"
1030
+ },
1031
+ "widgets_values": [
1032
+ "A veteran airmail pilot from a retro adventure film, expressed through the practical weathered flight costume. Realistic live-action costume character design, clean pale gray studio background, soft even light, natural material texture, relaxed upright stance with arms down, both hands and shoes fully visible, no text, logo, handheld props or additional people."
1033
+ ],
1034
+ "pos": [
1035
+ 400,
1036
+ 560
1037
+ ],
1038
+ "size": [
1039
+ 510,
1040
+ 210
1041
+ ],
1042
+ "title": "④ 造型 / Styling · Look, materials & pose"
1043
+ },
1044
+ {
1045
+ "id": 191,
1046
+ "type": "PreviewAny",
1047
+ "flags": {
1048
+ "collapsed": true
1049
+ },
1050
+ "mode": 0,
1051
+ "order": 20,
1052
+ "inputs": [
1053
+ {
1054
+ "name": "source",
1055
+ "type": "*",
1056
+ "link": 35
1057
+ }
1058
+ ],
1059
+ "outputs": [
1060
+ {
1061
+ "name": "STRING",
1062
+ "type": "STRING",
1063
+ "links": [],
1064
+ "slot_index": 0
1065
+ }
1066
+ ],
1067
+ "properties": {
1068
+ "Node name for S&R": "PreviewAny"
1069
+ },
1070
+ "widgets_values": [],
1071
+ "pos": [
1072
+ 20,
1073
+ 1580
1074
+ ],
1075
+ "size": [
1076
+ 850,
1077
+ 320
1078
+ ],
1079
+ "title": "本次区域条件 · JSON 核对"
1080
+ },
1081
+ {
1082
+ "id": 920,
1083
+ "type": "Note",
1084
+ "pos": [
1085
+ 20,
1086
+ 770
1087
+ ],
1088
+ "size": [
1089
+ 890,
1090
+ 230
1091
+ ],
1092
+ "flags": {},
1093
+ "order": 21,
1094
+ "mode": 0,
1095
+ "inputs": [],
1096
+ "outputs": [],
1097
+ "properties": {},
1098
+ "widgets_values": [
1099
+ "上传参考图 → 填写服装、身材、风格 → 选择尺寸 → 点击运行。\nUpload reference → Describe outfit, body & styling → Choose size → Run.\n随机服装、身材、造型或全部随机,填入后仍可手改。\nRandom outfit / body / styling / all fills editable text; image seed stays unchanged.\n固定三格:胸像 / 无头正面 / 完整后视。Three panels: portrait / headless front / full rear.\n支持中文与英文输入。Chinese and English prompts supported.\n表单需要 ComfyUI frontend ≥ 1.41.13;普通节点版使用 standard.json。\nApp Mode requires frontend ≥ 1.41.13. Use standard.json for the regular node graph.\n编码器默认 BF16,可自行选择兼容 FP8 / BF16 default; compatible FP8 is optional. 效果可能略有不同 / results may differ.\n安装 / Installation: see README.md."
1100
+ ],
1101
+ "title": "使用说明 / Quick start · candidate320"
1102
+ }
1103
+ ],
1104
+ "links": [
1105
+ [
1106
+ 1,
1107
+ 72,
1108
+ 0,
1109
+ 73,
1110
+ 0,
1111
+ "IMAGE"
1112
+ ],
1113
+ [
1114
+ 2,
1115
+ 57,
1116
+ 0,
1117
+ 73,
1118
+ 1,
1119
+ "VAE"
1120
+ ],
1121
+ [
1122
+ 3,
1123
+ 190,
1124
+ 2,
1125
+ 82,
1126
+ 0,
1127
+ "INT"
1128
+ ],
1129
+ [
1130
+ 4,
1131
+ 190,
1132
+ 3,
1133
+ 82,
1134
+ 1,
1135
+ "INT"
1136
+ ],
1137
+ [
1138
+ 5,
1139
+ 56,
1140
+ 0,
1141
+ 84,
1142
+ 0,
1143
+ "CLIP"
1144
+ ],
1145
+ [
1146
+ 6,
1147
+ 72,
1148
+ 0,
1149
+ 84,
1150
+ 1,
1151
+ "IMAGE"
1152
+ ],
1153
+ [
1154
+ 7,
1155
+ 190,
1156
+ 0,
1157
+ 84,
1158
+ 2,
1159
+ "STRING"
1160
+ ],
1161
+ [
1162
+ 8,
1163
+ 56,
1164
+ 0,
1165
+ 85,
1166
+ 0,
1167
+ "CLIP"
1168
+ ],
1169
+ [
1170
+ 9,
1171
+ 72,
1172
+ 0,
1173
+ 85,
1174
+ 1,
1175
+ "IMAGE"
1176
+ ],
1177
+ [
1178
+ 10,
1179
+ 55,
1180
+ 0,
1181
+ 79,
1182
+ 0,
1183
+ "MODEL"
1184
+ ],
1185
+ [
1186
+ 11,
1187
+ 73,
1188
+ 0,
1189
+ 79,
1190
+ 1,
1191
+ "LATENT"
1192
+ ],
1193
+ [
1194
+ 12,
1195
+ 57,
1196
+ 0,
1197
+ 79,
1198
+ 2,
1199
+ "VAE"
1200
+ ],
1201
+ [
1202
+ 13,
1203
+ 72,
1204
+ 0,
1205
+ 79,
1206
+ 3,
1207
+ "IMAGE"
1208
+ ],
1209
+ [
1210
+ 14,
1211
+ 82,
1212
+ 0,
1213
+ 79,
1214
+ 4,
1215
+ "LATENT"
1216
+ ],
1217
+ [
1218
+ 15,
1219
+ 180,
1220
+ 0,
1221
+ 53,
1222
+ 0,
1223
+ "MODEL"
1224
+ ],
1225
+ [
1226
+ 16,
1227
+ 180,
1228
+ 1,
1229
+ 53,
1230
+ 1,
1231
+ "CONDITIONING"
1232
+ ],
1233
+ [
1234
+ 17,
1235
+ 85,
1236
+ 0,
1237
+ 53,
1238
+ 2,
1239
+ "CONDITIONING"
1240
+ ],
1241
+ [
1242
+ 18,
1243
+ 82,
1244
+ 0,
1245
+ 53,
1246
+ 3,
1247
+ "LATENT"
1248
+ ],
1249
+ [
1250
+ 19,
1251
+ 120,
1252
+ 0,
1253
+ 53,
1254
+ 4,
1255
+ "SAMPLER"
1256
+ ],
1257
+ [
1258
+ 20,
1259
+ 121,
1260
+ 0,
1261
+ 53,
1262
+ 5,
1263
+ "SIGMAS"
1264
+ ],
1265
+ [
1266
+ 21,
1267
+ 190,
1268
+ 4,
1269
+ 53,
1270
+ 6,
1271
+ "INT"
1272
+ ],
1273
+ [
1274
+ 22,
1275
+ 181,
1276
+ 0,
1277
+ 54,
1278
+ 0,
1279
+ "LATENT"
1280
+ ],
1281
+ [
1282
+ 23,
1283
+ 57,
1284
+ 0,
1285
+ 54,
1286
+ 1,
1287
+ "VAE"
1288
+ ],
1289
+ [
1290
+ 24,
1291
+ 54,
1292
+ 0,
1293
+ 29,
1294
+ 0,
1295
+ "IMAGE"
1296
+ ],
1297
+ [
1298
+ 25,
1299
+ 79,
1300
+ 0,
1301
+ 180,
1302
+ 0,
1303
+ "MODEL"
1304
+ ],
1305
+ [
1306
+ 26,
1307
+ 56,
1308
+ 0,
1309
+ 180,
1310
+ 1,
1311
+ "CLIP"
1312
+ ],
1313
+ [
1314
+ 27,
1315
+ 84,
1316
+ 0,
1317
+ 180,
1318
+ 2,
1319
+ "CONDITIONING"
1320
+ ],
1321
+ [
1322
+ 28,
1323
+ 82,
1324
+ 0,
1325
+ 180,
1326
+ 3,
1327
+ "LATENT"
1328
+ ],
1329
+ [
1330
+ 29,
1331
+ 190,
1332
+ 1,
1333
+ 180,
1334
+ 4,
1335
+ "STRING"
1336
+ ],
1337
+ [
1338
+ 30,
1339
+ 53,
1340
+ 0,
1341
+ 181,
1342
+ 0,
1343
+ "LATENT"
1344
+ ],
1345
+ [
1346
+ 31,
1347
+ 180,
1348
+ 2,
1349
+ 181,
1350
+ 1,
1351
+ "GROUNDED_SPATIAL_STATE"
1352
+ ],
1353
+ [
1354
+ 32,
1355
+ 192,
1356
+ 0,
1357
+ 190,
1358
+ 0,
1359
+ "STRING"
1360
+ ],
1361
+ [
1362
+ 33,
1363
+ 193,
1364
+ 0,
1365
+ 190,
1366
+ 1,
1367
+ "STRING"
1368
+ ],
1369
+ [
1370
+ 34,
1371
+ 194,
1372
+ 0,
1373
+ 190,
1374
+ 2,
1375
+ "STRING"
1376
+ ],
1377
+ [
1378
+ 35,
1379
+ 190,
1380
+ 1,
1381
+ 191,
1382
+ 0,
1383
+ "STRING"
1384
+ ]
1385
+ ],
1386
+ "groups": [
1387
+ {
1388
+ "id": 1,
1389
+ "title": "三格人物卡 / Character Card · Inputs",
1390
+ "bounding": [
1391
+ 0,
1392
+ -40,
1393
+ 1660,
1394
+ 1080
1395
+ ],
1396
+ "color": "#3f789e",
1397
+ "font_size": 24,
1398
+ "flags": {}
1399
+ },
1400
+ {
1401
+ "id": 2,
1402
+ "title": "模型与采样 / Models & Sampling",
1403
+ "bounding": [
1404
+ 0,
1405
+ 1090,
1406
+ 1750,
1407
+ 750
1408
+ ],
1409
+ "color": "#59616b",
1410
+ "font_size": 24,
1411
+ "flags": {}
1412
+ }
1413
+ ],
1414
+ "config": {},
1415
+ "extra": {
1416
+ "ds": {
1417
+ "scale": 0.8,
1418
+ "offset": [
1419
+ 90,
1420
+ 170
1421
+ ]
1422
+ },
1423
+ "character_card_candidate": {
1424
+ "release_ready": false,
1425
+ "model_math": "shared numeric spatial branch; UI packaging only",
1426
+ "initial_adapter": "character-card-v04/candidate320-single.safetensors",
1427
+ "adapter_contract": {
1428
+ "adapter_path": "/home/coffmetoer/lora_train/storyflow/character-card/v0.4-rebuild/release_candidate320/candidate320-single.safetensors",
1429
+ "package": {
1430
+ "architecture_config_sha256": "65b85015e515bbfc3ffe6c03e10537a47121bd3756e0d21a1a87126da60bc4a8",
1431
+ "inputs": {
1432
+ "/home/coffmetoer/lora_train/archive/shared/00_shared_dependencies/Krea2-Identity-Edit/krea2_identity_edit_v1_2_r64.safetensors": "f794b47142555c929cf536a2f1e4f335174b9aedbb08572b07d45814d4242423",
1433
+ "/home/coffmetoer/lora_train/storyflow/character-card/v0.4-rebuild/config/grounded_formal90_compiled.json": "268b29b1ced9a2d9c68cbe7fa0ec1e67029418ca3756f6064342f33982dd5da5",
1434
+ "/home/coffmetoer/lora_train/storyflow/character-card/v0.4-rebuild/evaluation/formal320_paired_diagnostic/r002_biker_x034_076--1024x576--s20260910--measured/request.json": "ddb04f6fb0e3a05fce3a61146a6b98c381a1f4976a029a4b570a2ba0a8d0a9c6",
1435
+ "/home/coffmetoer/lora_train/storyflow/character-card/v0.4-rebuild/workspace/grounded_formal90_compiled/save/step-000000320.safetensors": "8f93f5f445e41e82d8426f3403908be35997177712bbc1f8ff6964bdb5ffa660"
1436
+ },
1437
+ "numerical_claim": "algebraically equivalent additive deltas; finite-precision runtime is not bitwise guaranteed",
1438
+ "output_lora_dtype": "float32",
1439
+ "protocol": "identity-task-lowrank-concat-v1",
1440
+ "rank_truncation": false,
1441
+ "remove_external_identity_loader": true,
1442
+ "spatial_source_sha256": "8f93f5f445e41e82d8426f3403908be35997177712bbc1f8ff6964bdb5ffa660",
1443
+ "status": "unselected_cpu_package_not_quality_approved",
1444
+ "tool_sha256": "47056b2f47aa7a42385168c43eb40a1302a22449b4aee874fabe26002a3c91ce",
1445
+ "unit_strengths": true
1446
+ },
1447
+ "identity_loading": "bundled_identity_external71_removed"
1448
+ },
1449
+ "initial_adapter_visual_status": "candidate320 66-image baseline; portrait expansion rejected after four-panel regression",
1450
+ "global_prompt_handling": "assembled clothing/body/styling and role template",
1451
+ "region_prompt_handling": "exact original CharacterCardAppearanceInputs baseline text; UI controls only",
1452
+ "source_UI_template": "/home/coffmetoer/lora_train/storyflow/character-card/v0.3-reference-edit/release-native/character_card_workflow.json",
1453
+ "single_reference": true,
1454
+ "batch_size": 1,
1455
+ "cfg": 1,
1456
+ "fixed_three_panels": true,
1457
+ "quality_fix_claimed": false
1458
+ },
1459
+ "linearMode": true,
1460
+ "linearData": {
1461
+ "inputs": [
1462
+ [
1463
+ "3579f887-385c-58eb-b676-e4c392bf0fd7:72:image",
1464
+ "参考图 / Reference",
1465
+ {
1466
+ "description": "参考图 / Reference · 单人且脸部清晰,半身或全身均可 / One person with a clear face; half/full-body photos are welcome.",
1467
+ "height": 180
1468
+ }
1469
+ ],
1470
+ [
1471
+ "3579f887-385c-58eb-b676-e4c392bf0fd7:192:value",
1472
+ "服装 / Outfit",
1473
+ {
1474
+ "description": "服装 / Outfit · 描述颜色、材质与搭配 / Describe colors, materials, layers and shoes. Chinese or English.",
1475
+ "height": 130
1476
+ }
1477
+ ],
1478
+ [
1479
+ "3579f887-385c-58eb-b676-e4c392bf0fd7:193:value",
1480
+ "身材 / Body",
1481
+ {
1482
+ "description": "身材 / Body · 描述肩宽、体态和比例 / Describe build and proportions.",
1483
+ "height": 90
1484
+ }
1485
+ ],
1486
+ [
1487
+ "3579f887-385c-58eb-b676-e4c392bf0fd7:194:value",
1488
+ "造型 / Styling",
1489
+ {
1490
+ "description": "造型 / Styling · 风格、材质与姿态 / Style, texture and pose; no need to describe the face.",
1491
+ "height": 90
1492
+ }
1493
+ ],
1494
+ [
1495
+ "3579f887-385c-58eb-b676-e4c392bf0fd7:190:resolution",
1496
+ "输出尺寸 / Resolution",
1497
+ {
1498
+ "description": "输出尺寸 / Resolution · 固定三格 / Three panels, 16:9."
1499
+ }
1500
+ ],
1501
+ [
1502
+ "3579f887-385c-58eb-b676-e4c392bf0fd7:190:seed",
1503
+ "种子 / Seed",
1504
+ {
1505
+ "description": "种子 / Seed · 保持相同输入以复现 / Keep inputs and seed to reproduce."
1506
+ }
1507
+ ]
1508
+ ],
1509
+ "outputs": [
1510
+ 29
1511
+ ]
1512
+ }
1513
+ },
1514
+ "version": 0.4
1515
+ }
storyflow/character-card-v0.4/Storyflow-v04-spatial-node-encoder-fix.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e4ad3f584886e871015014a4156d01633578fe12df974fa170ac15407171604
3
+ size 25521
storyflow/character-card-v0.4/Storyflow-v04-standard-BF16-default.json ADDED
@@ -0,0 +1,1462 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "id": "3579f887-385c-58eb-b676-e4c392bf0fd7",
3
+ "revision": 0,
4
+ "last_node_id": 920,
5
+ "last_link_id": 35,
6
+ "nodes": [
7
+ {
8
+ "id": 55,
9
+ "type": "UNETLoader",
10
+ "pos": [
11
+ 20,
12
+ 1180
13
+ ],
14
+ "size": [
15
+ 360,
16
+ 130
17
+ ],
18
+ "flags": {
19
+ "collapsed": true
20
+ },
21
+ "order": 0,
22
+ "mode": 0,
23
+ "inputs": [],
24
+ "outputs": [
25
+ {
26
+ "name": "MODEL",
27
+ "type": "MODEL",
28
+ "links": [
29
+ 10
30
+ ],
31
+ "slot_index": 0
32
+ }
33
+ ],
34
+ "properties": {
35
+ "Node name for S&R": "UNETLoader"
36
+ },
37
+ "widgets_values": [
38
+ "krea2_turbo_bf16.safetensors",
39
+ "default"
40
+ ],
41
+ "title": "Krea2 模型"
42
+ },
43
+ {
44
+ "id": 56,
45
+ "type": "CharacterCardGroundedTextEncoder",
46
+ "pos": [
47
+ 360,
48
+ 1180
49
+ ],
50
+ "size": [
51
+ 470,
52
+ 130
53
+ ],
54
+ "flags": {
55
+ "collapsed": true
56
+ },
57
+ "order": 1,
58
+ "mode": 0,
59
+ "inputs": [],
60
+ "outputs": [
61
+ {
62
+ "name": "clip",
63
+ "type": "CLIP",
64
+ "links": [
65
+ 5,
66
+ 8,
67
+ 26
68
+ ],
69
+ "slot_index": 0
70
+ }
71
+ ],
72
+ "properties": {
73
+ "Node name for S&R": "CharacterCardGroundedTextEncoder"
74
+ },
75
+ "widgets_values": [
76
+ "character-card/qwen3vl_training_bf16.safetensors"
77
+ ],
78
+ "title": "文本编码器 / Text encoder · BF16 default, FP8 optional"
79
+ },
80
+ {
81
+ "id": 57,
82
+ "type": "VAELoader",
83
+ "pos": [
84
+ 700,
85
+ 1180
86
+ ],
87
+ "size": [
88
+ 390,
89
+ 130
90
+ ],
91
+ "flags": {
92
+ "collapsed": true
93
+ },
94
+ "order": 2,
95
+ "mode": 0,
96
+ "inputs": [],
97
+ "outputs": [
98
+ {
99
+ "name": "VAE",
100
+ "type": "VAE",
101
+ "links": [
102
+ 2,
103
+ 12,
104
+ 23
105
+ ],
106
+ "slot_index": 0
107
+ }
108
+ ],
109
+ "properties": {
110
+ "Node name for S&R": "VAELoader"
111
+ },
112
+ "widgets_values": [
113
+ "qwen/qwen_image_vae.safetensors"
114
+ ],
115
+ "title": "VAE"
116
+ },
117
+ {
118
+ "id": 72,
119
+ "type": "LoadImage",
120
+ "pos": [
121
+ 20,
122
+ 20
123
+ ],
124
+ "size": [
125
+ 330,
126
+ 430
127
+ ],
128
+ "flags": {},
129
+ "order": 3,
130
+ "mode": 0,
131
+ "inputs": [
132
+ {
133
+ "name": "image",
134
+ "type": "COMBO",
135
+ "link": null,
136
+ "widget": {
137
+ "name": "image"
138
+ },
139
+ "label": "参考图 / Reference"
140
+ }
141
+ ],
142
+ "outputs": [
143
+ {
144
+ "name": "IMAGE",
145
+ "type": "IMAGE",
146
+ "links": [
147
+ 1,
148
+ 6,
149
+ 9,
150
+ 13
151
+ ],
152
+ "slot_index": 0
153
+ },
154
+ {
155
+ "name": "MASK",
156
+ "type": "MASK",
157
+ "links": [],
158
+ "slot_index": 1
159
+ }
160
+ ],
161
+ "properties": {
162
+ "Node name for S&R": "LoadImage"
163
+ },
164
+ "widgets_values": [
165
+ "storyflow-v03/df34f46583e7d1ebf445e93db03c15a533155c096075a043cf351a42d8fcb181.png",
166
+ "image"
167
+ ],
168
+ "title": "① 参考图 / Reference · Upload photo"
169
+ },
170
+ {
171
+ "id": 73,
172
+ "type": "VAEEncode",
173
+ "pos": [
174
+ 1380,
175
+ 1180
176
+ ],
177
+ "size": [
178
+ 340,
179
+ 120
180
+ ],
181
+ "flags": {
182
+ "collapsed": true
183
+ },
184
+ "order": 4,
185
+ "mode": 0,
186
+ "inputs": [
187
+ {
188
+ "name": "pixels",
189
+ "type": "IMAGE",
190
+ "link": 1
191
+ },
192
+ {
193
+ "name": "vae",
194
+ "type": "VAE",
195
+ "link": 2
196
+ }
197
+ ],
198
+ "outputs": [
199
+ {
200
+ "name": "LATENT",
201
+ "type": "LATENT",
202
+ "links": [
203
+ 11
204
+ ],
205
+ "slot_index": 0
206
+ }
207
+ ],
208
+ "properties": {
209
+ "Node name for S&R": "VAEEncode"
210
+ },
211
+ "widgets_values": [],
212
+ "title": "参考 VAE 编码"
213
+ },
214
+ {
215
+ "id": 82,
216
+ "type": "EmptySD3LatentImage",
217
+ "pos": [
218
+ 20,
219
+ 1280
220
+ ],
221
+ "size": [
222
+ 370,
223
+ 150
224
+ ],
225
+ "flags": {
226
+ "collapsed": true
227
+ },
228
+ "order": 9,
229
+ "mode": 0,
230
+ "inputs": [
231
+ {
232
+ "name": "width",
233
+ "type": "INT",
234
+ "link": 3,
235
+ "widget": {
236
+ "name": "width"
237
+ }
238
+ },
239
+ {
240
+ "name": "height",
241
+ "type": "INT",
242
+ "link": 4,
243
+ "widget": {
244
+ "name": "height"
245
+ }
246
+ }
247
+ ],
248
+ "outputs": [
249
+ {
250
+ "name": "LATENT",
251
+ "type": "LATENT",
252
+ "links": [
253
+ 14,
254
+ 18,
255
+ 28
256
+ ],
257
+ "slot_index": 0
258
+ }
259
+ ],
260
+ "properties": {
261
+ "Node name for S&R": "EmptySD3LatentImage"
262
+ },
263
+ "widgets_values": [
264
+ 1024,
265
+ 576,
266
+ 1
267
+ ],
268
+ "title": "③ 画布 · width / height"
269
+ },
270
+ {
271
+ "id": 84,
272
+ "type": "Krea2EditGroundedEncode",
273
+ "pos": [
274
+ 360,
275
+ 1280
276
+ ],
277
+ "size": [
278
+ 460,
279
+ 260
280
+ ],
281
+ "flags": {
282
+ "collapsed": true
283
+ },
284
+ "order": 10,
285
+ "mode": 0,
286
+ "inputs": [
287
+ {
288
+ "name": "clip",
289
+ "type": "CLIP",
290
+ "link": 5
291
+ },
292
+ {
293
+ "name": "image",
294
+ "type": "IMAGE",
295
+ "link": 6
296
+ },
297
+ {
298
+ "name": "prompt",
299
+ "type": "STRING",
300
+ "link": 7,
301
+ "widget": {
302
+ "name": "prompt"
303
+ }
304
+ }
305
+ ],
306
+ "outputs": [
307
+ {
308
+ "name": "CONDITIONING",
309
+ "type": "CONDITIONING",
310
+ "links": [
311
+ 27
312
+ ],
313
+ "slot_index": 0
314
+ }
315
+ ],
316
+ "properties": {
317
+ "Node name for S&R": "Krea2EditGroundedEncode"
318
+ },
319
+ "widgets_values": [
320
+ "",
321
+ 768,
322
+ ""
323
+ ],
324
+ "title": "全局描述 + 同一参考图"
325
+ },
326
+ {
327
+ "id": 85,
328
+ "type": "Krea2EditGroundedEncode",
329
+ "pos": [
330
+ 700,
331
+ 1280
332
+ ],
333
+ "size": [
334
+ 460,
335
+ 260
336
+ ],
337
+ "flags": {
338
+ "collapsed": true
339
+ },
340
+ "order": 11,
341
+ "mode": 0,
342
+ "inputs": [
343
+ {
344
+ "name": "clip",
345
+ "type": "CLIP",
346
+ "link": 8
347
+ },
348
+ {
349
+ "name": "image",
350
+ "type": "IMAGE",
351
+ "link": 9
352
+ }
353
+ ],
354
+ "outputs": [
355
+ {
356
+ "name": "CONDITIONING",
357
+ "type": "CONDITIONING",
358
+ "links": [
359
+ 17
360
+ ],
361
+ "slot_index": 0
362
+ }
363
+ ],
364
+ "title": "空负向 · CFG 1",
365
+ "properties": {
366
+ "Node name for S&R": "Krea2EditGroundedEncode"
367
+ },
368
+ "widgets_values": [
369
+ "",
370
+ 768,
371
+ ""
372
+ ]
373
+ },
374
+ {
375
+ "id": 79,
376
+ "type": "Krea2EditModelPatch",
377
+ "pos": [
378
+ 1040,
379
+ 1280
380
+ ],
381
+ "size": [
382
+ 390,
383
+ 260
384
+ ],
385
+ "flags": {
386
+ "collapsed": true
387
+ },
388
+ "order": 12,
389
+ "mode": 0,
390
+ "inputs": [
391
+ {
392
+ "name": "model",
393
+ "type": "MODEL",
394
+ "link": 10
395
+ },
396
+ {
397
+ "name": "source_latent",
398
+ "type": "LATENT",
399
+ "link": 11
400
+ },
401
+ {
402
+ "name": "vae",
403
+ "type": "VAE",
404
+ "link": 12
405
+ },
406
+ {
407
+ "name": "source_image",
408
+ "type": "IMAGE",
409
+ "link": 13
410
+ },
411
+ {
412
+ "name": "target_latent",
413
+ "type": "LATENT",
414
+ "link": 14
415
+ }
416
+ ],
417
+ "outputs": [
418
+ {
419
+ "name": "MODEL",
420
+ "type": "MODEL",
421
+ "links": [
422
+ 25
423
+ ],
424
+ "slot_index": 0
425
+ }
426
+ ],
427
+ "properties": {
428
+ "Node name for S&R": "Krea2EditModelPatch"
429
+ },
430
+ "widgets_values": [
431
+ 1.0,
432
+ 1.0,
433
+ "fit"
434
+ ],
435
+ "title": "参考条件 · fit / strength 1"
436
+ },
437
+ {
438
+ "id": 53,
439
+ "type": "SamplerCustom",
440
+ "pos": [
441
+ 1380,
442
+ 1280
443
+ ],
444
+ "size": [
445
+ 370,
446
+ 330
447
+ ],
448
+ "flags": {
449
+ "collapsed": true
450
+ },
451
+ "order": 16,
452
+ "mode": 0,
453
+ "inputs": [
454
+ {
455
+ "name": "model",
456
+ "type": "MODEL",
457
+ "link": 15
458
+ },
459
+ {
460
+ "name": "positive",
461
+ "type": "CONDITIONING",
462
+ "link": 16
463
+ },
464
+ {
465
+ "name": "negative",
466
+ "type": "CONDITIONING",
467
+ "link": 17
468
+ },
469
+ {
470
+ "name": "latent_image",
471
+ "type": "LATENT",
472
+ "link": 18
473
+ },
474
+ {
475
+ "name": "sampler",
476
+ "type": "SAMPLER",
477
+ "link": 19
478
+ },
479
+ {
480
+ "name": "sigmas",
481
+ "type": "SIGMAS",
482
+ "link": 20
483
+ },
484
+ {
485
+ "name": "noise_seed",
486
+ "type": "INT",
487
+ "link": 21,
488
+ "widget": {
489
+ "name": "noise_seed"
490
+ }
491
+ }
492
+ ],
493
+ "outputs": [
494
+ {
495
+ "name": "output",
496
+ "type": "LATENT",
497
+ "links": [
498
+ 30
499
+ ],
500
+ "slot_index": 0
501
+ },
502
+ {
503
+ "name": "denoised_output",
504
+ "type": "LATENT",
505
+ "links": [],
506
+ "slot_index": 1
507
+ }
508
+ ],
509
+ "properties": {
510
+ "Node name for S&R": "SamplerCustom"
511
+ },
512
+ "widgets_values": [
513
+ true,
514
+ 20260910,
515
+ "fixed",
516
+ 1.0
517
+ ],
518
+ "title": "④ 种子 · fixed / CFG 1"
519
+ },
520
+ {
521
+ "id": 54,
522
+ "type": "VAEDecode",
523
+ "pos": [
524
+ 20,
525
+ 1380
526
+ ],
527
+ "size": [
528
+ 330,
529
+ 150
530
+ ],
531
+ "flags": {
532
+ "collapsed": true
533
+ },
534
+ "order": 18,
535
+ "mode": 0,
536
+ "inputs": [
537
+ {
538
+ "name": "samples",
539
+ "type": "LATENT",
540
+ "link": 22
541
+ },
542
+ {
543
+ "name": "vae",
544
+ "type": "VAE",
545
+ "link": 23
546
+ }
547
+ ],
548
+ "outputs": [
549
+ {
550
+ "name": "IMAGE",
551
+ "type": "IMAGE",
552
+ "links": [
553
+ 24
554
+ ],
555
+ "slot_index": 0
556
+ }
557
+ ],
558
+ "properties": {
559
+ "Node name for S&R": "VAEDecode"
560
+ },
561
+ "widgets_values": [],
562
+ "title": "原画布解码"
563
+ },
564
+ {
565
+ "id": 29,
566
+ "type": "SaveImage",
567
+ "pos": [
568
+ 950,
569
+ 20
570
+ ],
571
+ "size": [
572
+ 670,
573
+ 790
574
+ ],
575
+ "flags": {},
576
+ "order": 19,
577
+ "mode": 0,
578
+ "inputs": [
579
+ {
580
+ "name": "images",
581
+ "type": "IMAGE",
582
+ "link": 24
583
+ }
584
+ ],
585
+ "outputs": [
586
+ {
587
+ "name": "images",
588
+ "type": "IMAGE",
589
+ "links": [],
590
+ "slot_index": 0
591
+ }
592
+ ],
593
+ "properties": {
594
+ "Node name for S&R": "SaveImage"
595
+ },
596
+ "widgets_values": [
597
+ "storyflow-v04/candidate_native_card"
598
+ ],
599
+ "title": "⑥ 人物卡 / Character card · Preview & Save"
600
+ },
601
+ {
602
+ "id": 120,
603
+ "pos": [
604
+ 700,
605
+ 1380
606
+ ],
607
+ "size": [
608
+ 360,
609
+ 120
610
+ ],
611
+ "flags": {
612
+ "collapsed": true
613
+ },
614
+ "properties": {
615
+ "Node name for S&R": "KSamplerSelect"
616
+ },
617
+ "title": "Euler",
618
+ "type": "KSamplerSelect",
619
+ "order": 14,
620
+ "mode": 0,
621
+ "inputs": [],
622
+ "outputs": [
623
+ {
624
+ "name": "SAMPLER",
625
+ "type": "SAMPLER",
626
+ "links": [
627
+ 19
628
+ ],
629
+ "slot_index": 0
630
+ }
631
+ ],
632
+ "widgets_values": [
633
+ "euler"
634
+ ]
635
+ },
636
+ {
637
+ "id": 121,
638
+ "pos": [
639
+ 1040,
640
+ 1380
641
+ ],
642
+ "size": [
643
+ 470,
644
+ 220
645
+ ],
646
+ "flags": {
647
+ "collapsed": true
648
+ },
649
+ "properties": {
650
+ "Node name for S&R": "ManualSigmas"
651
+ },
652
+ "title": "固定 8 步 sigmas",
653
+ "type": "ManualSigmas",
654
+ "order": 15,
655
+ "mode": 0,
656
+ "inputs": [],
657
+ "outputs": [
658
+ {
659
+ "name": "SIGMAS",
660
+ "type": "SIGMAS",
661
+ "links": [
662
+ 20
663
+ ],
664
+ "slot_index": 0
665
+ }
666
+ ],
667
+ "widgets_values": [
668
+ "1.0,0.9248116147173576,0.8367657420590665,0.7322560154608024,0.6061882408399552,0.4511248963112157,0.25576263215491235,0.0020454536906595946,0.0"
669
+ ]
670
+ },
671
+ {
672
+ "id": 180,
673
+ "type": "CharacterCardGroundedSpatial",
674
+ "flags": {
675
+ "collapsed": true
676
+ },
677
+ "mode": 0,
678
+ "order": 13,
679
+ "inputs": [
680
+ {
681
+ "name": "model",
682
+ "type": "MODEL",
683
+ "link": 25
684
+ },
685
+ {
686
+ "name": "clip",
687
+ "type": "CLIP",
688
+ "link": 26
689
+ },
690
+ {
691
+ "name": "global_grounded",
692
+ "type": "CONDITIONING",
693
+ "link": 27
694
+ },
695
+ {
696
+ "name": "target_latent",
697
+ "type": "LATENT",
698
+ "link": 28
699
+ },
700
+ {
701
+ "name": "regions_json",
702
+ "type": "STRING",
703
+ "link": 29,
704
+ "widget": {
705
+ "name": "regions_json"
706
+ }
707
+ }
708
+ ],
709
+ "outputs": [
710
+ {
711
+ "name": "model",
712
+ "type": "MODEL",
713
+ "links": [
714
+ 15
715
+ ],
716
+ "slot_index": 0
717
+ },
718
+ {
719
+ "name": "conditioning",
720
+ "type": "CONDITIONING",
721
+ "links": [
722
+ 16
723
+ ],
724
+ "slot_index": 1
725
+ },
726
+ {
727
+ "name": "state",
728
+ "type": "GROUNDED_SPATIAL_STATE",
729
+ "links": [
730
+ 31
731
+ ],
732
+ "slot_index": 2
733
+ }
734
+ ],
735
+ "properties": {
736
+ "Node name for S&R": "CharacterCardGroundedSpatial"
737
+ },
738
+ "widgets_values": [
739
+ "character-card-v04/candidate320-single.safetensors",
740
+ ""
741
+ ],
742
+ "pos": [
743
+ 1380,
744
+ 1380
745
+ ],
746
+ "size": [
747
+ 850,
748
+ 360
749
+ ],
750
+ "title": "候选空间权重 · 选择 task + spatial 单文件"
751
+ },
752
+ {
753
+ "id": 181,
754
+ "type": "CharacterCardGroundedSpatialReport",
755
+ "flags": {
756
+ "collapsed": true
757
+ },
758
+ "mode": 0,
759
+ "order": 17,
760
+ "inputs": [
761
+ {
762
+ "name": "samples",
763
+ "type": "LATENT",
764
+ "link": 30
765
+ },
766
+ {
767
+ "name": "state",
768
+ "type": "GROUNDED_SPATIAL_STATE",
769
+ "link": 31
770
+ }
771
+ ],
772
+ "outputs": [
773
+ {
774
+ "name": "samples",
775
+ "type": "LATENT",
776
+ "links": [
777
+ 22
778
+ ],
779
+ "slot_index": 0
780
+ },
781
+ {
782
+ "name": "report",
783
+ "type": "STRING",
784
+ "links": [],
785
+ "slot_index": 1
786
+ }
787
+ ],
788
+ "properties": {
789
+ "Node name for S&R": "CharacterCardGroundedSpatialReport"
790
+ },
791
+ "widgets_values": [],
792
+ "pos": [
793
+ 20,
794
+ 1480
795
+ ],
796
+ "size": [
797
+ 330,
798
+ 180
799
+ ],
800
+ "title": "空间执行报告"
801
+ },
802
+ {
803
+ "id": 190,
804
+ "type": "CharacterCardFixedAppearanceInputs",
805
+ "flags": {},
806
+ "mode": 0,
807
+ "order": 8,
808
+ "inputs": [
809
+ {
810
+ "name": "outfit",
811
+ "type": "STRING",
812
+ "link": 32,
813
+ "widget": {
814
+ "name": "outfit"
815
+ }
816
+ },
817
+ {
818
+ "name": "body",
819
+ "type": "STRING",
820
+ "link": 33,
821
+ "widget": {
822
+ "name": "body"
823
+ }
824
+ },
825
+ {
826
+ "name": "styling",
827
+ "type": "STRING",
828
+ "link": 34,
829
+ "widget": {
830
+ "name": "styling"
831
+ }
832
+ },
833
+ {
834
+ "name": "resolution",
835
+ "type": "COMBO",
836
+ "link": null,
837
+ "widget": {
838
+ "name": "resolution"
839
+ },
840
+ "label": "输出尺寸 / Resolution"
841
+ },
842
+ {
843
+ "name": "seed",
844
+ "type": "INT",
845
+ "link": null,
846
+ "widget": {
847
+ "name": "seed"
848
+ },
849
+ "label": "种子 / Seed"
850
+ }
851
+ ],
852
+ "outputs": [
853
+ {
854
+ "name": "global_prompt",
855
+ "type": "STRING",
856
+ "links": [
857
+ 7
858
+ ],
859
+ "slot_index": 0
860
+ },
861
+ {
862
+ "name": "regions_json",
863
+ "type": "STRING",
864
+ "links": [
865
+ 29,
866
+ 35
867
+ ],
868
+ "slot_index": 1
869
+ },
870
+ {
871
+ "name": "width",
872
+ "type": "INT",
873
+ "links": [
874
+ 3
875
+ ],
876
+ "slot_index": 2
877
+ },
878
+ {
879
+ "name": "height",
880
+ "type": "INT",
881
+ "links": [
882
+ 4
883
+ ],
884
+ "slot_index": 3
885
+ },
886
+ {
887
+ "name": "seed",
888
+ "type": "INT",
889
+ "links": [
890
+ 21
891
+ ],
892
+ "slot_index": 4
893
+ }
894
+ ],
895
+ "properties": {
896
+ "Node name for S&R": "CharacterCardFixedAppearanceInputs"
897
+ },
898
+ "widgets_values": [
899
+ "",
900
+ "",
901
+ "",
902
+ "1024x576",
903
+ 20260910
904
+ ],
905
+ "pos": [
906
+ 20,
907
+ 520
908
+ ],
909
+ "size": [
910
+ 330,
911
+ 180
912
+ ],
913
+ "title": "⑤ 尺寸与种子 / Size & Seed"
914
+ },
915
+ {
916
+ "id": 192,
917
+ "type": "PrimitiveStringMultiline",
918
+ "flags": {},
919
+ "mode": 0,
920
+ "order": 5,
921
+ "inputs": [
922
+ {
923
+ "name": "value",
924
+ "type": "STRING",
925
+ "link": null,
926
+ "widget": {
927
+ "name": "value"
928
+ },
929
+ "label": "服装 / Outfit"
930
+ }
931
+ ],
932
+ "outputs": [
933
+ {
934
+ "name": "STRING",
935
+ "type": "STRING",
936
+ "links": [
937
+ 32
938
+ ],
939
+ "slot_index": 0
940
+ }
941
+ ],
942
+ "properties": {
943
+ "Node name for S&R": "PrimitiveStringMultiline"
944
+ },
945
+ "widgets_values": [
946
+ "An ochre tan cropped leather flight jacket with an ivory shearling collar, brass front zipper, ribbed cuffs and two angled flap pockets; an ivory ribbed turtleneck underneath; dark brown straight cargo trousers with modest thigh pockets and brown lace-up leather ankle boots. The back of the jacket has a plain horizontal shoulder yoke and a neat waist seam, no lettering."
947
+ ],
948
+ "pos": [
949
+ 400,
950
+ 20
951
+ ],
952
+ "size": [
953
+ 510,
954
+ 210
955
+ ],
956
+ "title": "② 服装 / Outfit · Colors, layers & shoes"
957
+ },
958
+ {
959
+ "id": 193,
960
+ "type": "PrimitiveStringMultiline",
961
+ "flags": {},
962
+ "mode": 0,
963
+ "order": 6,
964
+ "inputs": [
965
+ {
966
+ "name": "value",
967
+ "type": "STRING",
968
+ "link": null,
969
+ "widget": {
970
+ "name": "value"
971
+ },
972
+ "label": "身材 / Body"
973
+ }
974
+ ],
975
+ "outputs": [
976
+ {
977
+ "name": "STRING",
978
+ "type": "STRING",
979
+ "links": [
980
+ 33
981
+ ],
982
+ "slot_index": 0
983
+ }
984
+ ],
985
+ "properties": {
986
+ "Node name for S&R": "PrimitiveStringMultiline"
987
+ },
988
+ "widgets_values": [
989
+ "The person has a compact athletic build, sturdy shoulders, fit torso and naturally strong arms and legs, balanced realistic proportions."
990
+ ],
991
+ "pos": [
992
+ 400,
993
+ 290
994
+ ],
995
+ "size": [
996
+ 510,
997
+ 210
998
+ ],
999
+ "title": "③ 身材 / Body · Free-form description"
1000
+ },
1001
+ {
1002
+ "id": 194,
1003
+ "type": "PrimitiveStringMultiline",
1004
+ "flags": {},
1005
+ "mode": 0,
1006
+ "order": 7,
1007
+ "inputs": [
1008
+ {
1009
+ "name": "value",
1010
+ "type": "STRING",
1011
+ "link": null,
1012
+ "widget": {
1013
+ "name": "value"
1014
+ },
1015
+ "label": "风格 / Styling"
1016
+ }
1017
+ ],
1018
+ "outputs": [
1019
+ {
1020
+ "name": "STRING",
1021
+ "type": "STRING",
1022
+ "links": [
1023
+ 34
1024
+ ],
1025
+ "slot_index": 0
1026
+ }
1027
+ ],
1028
+ "properties": {
1029
+ "Node name for S&R": "PrimitiveStringMultiline"
1030
+ },
1031
+ "widgets_values": [
1032
+ "A veteran airmail pilot from a retro adventure film, expressed through the practical weathered flight costume. Realistic live-action costume character design, clean pale gray studio background, soft even light, natural material texture, relaxed upright stance with arms down, both hands and shoes fully visible, no text, logo, handheld props or additional people."
1033
+ ],
1034
+ "pos": [
1035
+ 400,
1036
+ 560
1037
+ ],
1038
+ "size": [
1039
+ 510,
1040
+ 210
1041
+ ],
1042
+ "title": "④ 造型 / Styling · Look, materials & pose"
1043
+ },
1044
+ {
1045
+ "id": 191,
1046
+ "type": "PreviewAny",
1047
+ "flags": {
1048
+ "collapsed": true
1049
+ },
1050
+ "mode": 0,
1051
+ "order": 20,
1052
+ "inputs": [
1053
+ {
1054
+ "name": "source",
1055
+ "type": "*",
1056
+ "link": 35
1057
+ }
1058
+ ],
1059
+ "outputs": [
1060
+ {
1061
+ "name": "STRING",
1062
+ "type": "STRING",
1063
+ "links": [],
1064
+ "slot_index": 0
1065
+ }
1066
+ ],
1067
+ "properties": {
1068
+ "Node name for S&R": "PreviewAny"
1069
+ },
1070
+ "widgets_values": [],
1071
+ "pos": [
1072
+ 20,
1073
+ 1580
1074
+ ],
1075
+ "size": [
1076
+ 850,
1077
+ 320
1078
+ ],
1079
+ "title": "本次区域条件 · JSON 核对"
1080
+ },
1081
+ {
1082
+ "id": 920,
1083
+ "type": "Note",
1084
+ "pos": [
1085
+ 20,
1086
+ 770
1087
+ ],
1088
+ "size": [
1089
+ 890,
1090
+ 230
1091
+ ],
1092
+ "flags": {},
1093
+ "order": 21,
1094
+ "mode": 0,
1095
+ "inputs": [],
1096
+ "outputs": [],
1097
+ "properties": {},
1098
+ "widgets_values": [
1099
+ "上传参考图 → 填写服装、身材、风格 → 选择尺寸 → 点击运行。\nUpload reference → Describe outfit, body & styling → Choose size → Run.\n随机服装、身材、造型或全部随机,填入后仍可手改。\nRandom outfit / body / styling / all fills editable text; image seed stays unchanged.\n固定三格:胸像 / 无头正面 / 完整后视。Three panels: portrait / headless front / full rear.\n支持中文与英文输入。Chinese and English prompts supported.\n表单需要 ComfyUI frontend ≥ 1.41.13;普通节点版使用 standard.json。\nApp Mode requires frontend ≥ 1.41.13. Use standard.json for the regular node graph.\n编码器默认 BF16,可自行选择兼容 FP8 / BF16 default; compatible FP8 is optional. 效果可能略有不同 / results may differ.\n安装 / Installation: see README.md."
1100
+ ],
1101
+ "title": "使用说明 / Quick start · candidate320"
1102
+ }
1103
+ ],
1104
+ "links": [
1105
+ [
1106
+ 1,
1107
+ 72,
1108
+ 0,
1109
+ 73,
1110
+ 0,
1111
+ "IMAGE"
1112
+ ],
1113
+ [
1114
+ 2,
1115
+ 57,
1116
+ 0,
1117
+ 73,
1118
+ 1,
1119
+ "VAE"
1120
+ ],
1121
+ [
1122
+ 3,
1123
+ 190,
1124
+ 2,
1125
+ 82,
1126
+ 0,
1127
+ "INT"
1128
+ ],
1129
+ [
1130
+ 4,
1131
+ 190,
1132
+ 3,
1133
+ 82,
1134
+ 1,
1135
+ "INT"
1136
+ ],
1137
+ [
1138
+ 5,
1139
+ 56,
1140
+ 0,
1141
+ 84,
1142
+ 0,
1143
+ "CLIP"
1144
+ ],
1145
+ [
1146
+ 6,
1147
+ 72,
1148
+ 0,
1149
+ 84,
1150
+ 1,
1151
+ "IMAGE"
1152
+ ],
1153
+ [
1154
+ 7,
1155
+ 190,
1156
+ 0,
1157
+ 84,
1158
+ 2,
1159
+ "STRING"
1160
+ ],
1161
+ [
1162
+ 8,
1163
+ 56,
1164
+ 0,
1165
+ 85,
1166
+ 0,
1167
+ "CLIP"
1168
+ ],
1169
+ [
1170
+ 9,
1171
+ 72,
1172
+ 0,
1173
+ 85,
1174
+ 1,
1175
+ "IMAGE"
1176
+ ],
1177
+ [
1178
+ 10,
1179
+ 55,
1180
+ 0,
1181
+ 79,
1182
+ 0,
1183
+ "MODEL"
1184
+ ],
1185
+ [
1186
+ 11,
1187
+ 73,
1188
+ 0,
1189
+ 79,
1190
+ 1,
1191
+ "LATENT"
1192
+ ],
1193
+ [
1194
+ 12,
1195
+ 57,
1196
+ 0,
1197
+ 79,
1198
+ 2,
1199
+ "VAE"
1200
+ ],
1201
+ [
1202
+ 13,
1203
+ 72,
1204
+ 0,
1205
+ 79,
1206
+ 3,
1207
+ "IMAGE"
1208
+ ],
1209
+ [
1210
+ 14,
1211
+ 82,
1212
+ 0,
1213
+ 79,
1214
+ 4,
1215
+ "LATENT"
1216
+ ],
1217
+ [
1218
+ 15,
1219
+ 180,
1220
+ 0,
1221
+ 53,
1222
+ 0,
1223
+ "MODEL"
1224
+ ],
1225
+ [
1226
+ 16,
1227
+ 180,
1228
+ 1,
1229
+ 53,
1230
+ 1,
1231
+ "CONDITIONING"
1232
+ ],
1233
+ [
1234
+ 17,
1235
+ 85,
1236
+ 0,
1237
+ 53,
1238
+ 2,
1239
+ "CONDITIONING"
1240
+ ],
1241
+ [
1242
+ 18,
1243
+ 82,
1244
+ 0,
1245
+ 53,
1246
+ 3,
1247
+ "LATENT"
1248
+ ],
1249
+ [
1250
+ 19,
1251
+ 120,
1252
+ 0,
1253
+ 53,
1254
+ 4,
1255
+ "SAMPLER"
1256
+ ],
1257
+ [
1258
+ 20,
1259
+ 121,
1260
+ 0,
1261
+ 53,
1262
+ 5,
1263
+ "SIGMAS"
1264
+ ],
1265
+ [
1266
+ 21,
1267
+ 190,
1268
+ 4,
1269
+ 53,
1270
+ 6,
1271
+ "INT"
1272
+ ],
1273
+ [
1274
+ 22,
1275
+ 181,
1276
+ 0,
1277
+ 54,
1278
+ 0,
1279
+ "LATENT"
1280
+ ],
1281
+ [
1282
+ 23,
1283
+ 57,
1284
+ 0,
1285
+ 54,
1286
+ 1,
1287
+ "VAE"
1288
+ ],
1289
+ [
1290
+ 24,
1291
+ 54,
1292
+ 0,
1293
+ 29,
1294
+ 0,
1295
+ "IMAGE"
1296
+ ],
1297
+ [
1298
+ 25,
1299
+ 79,
1300
+ 0,
1301
+ 180,
1302
+ 0,
1303
+ "MODEL"
1304
+ ],
1305
+ [
1306
+ 26,
1307
+ 56,
1308
+ 0,
1309
+ 180,
1310
+ 1,
1311
+ "CLIP"
1312
+ ],
1313
+ [
1314
+ 27,
1315
+ 84,
1316
+ 0,
1317
+ 180,
1318
+ 2,
1319
+ "CONDITIONING"
1320
+ ],
1321
+ [
1322
+ 28,
1323
+ 82,
1324
+ 0,
1325
+ 180,
1326
+ 3,
1327
+ "LATENT"
1328
+ ],
1329
+ [
1330
+ 29,
1331
+ 190,
1332
+ 1,
1333
+ 180,
1334
+ 4,
1335
+ "STRING"
1336
+ ],
1337
+ [
1338
+ 30,
1339
+ 53,
1340
+ 0,
1341
+ 181,
1342
+ 0,
1343
+ "LATENT"
1344
+ ],
1345
+ [
1346
+ 31,
1347
+ 180,
1348
+ 2,
1349
+ 181,
1350
+ 1,
1351
+ "GROUNDED_SPATIAL_STATE"
1352
+ ],
1353
+ [
1354
+ 32,
1355
+ 192,
1356
+ 0,
1357
+ 190,
1358
+ 0,
1359
+ "STRING"
1360
+ ],
1361
+ [
1362
+ 33,
1363
+ 193,
1364
+ 0,
1365
+ 190,
1366
+ 1,
1367
+ "STRING"
1368
+ ],
1369
+ [
1370
+ 34,
1371
+ 194,
1372
+ 0,
1373
+ 190,
1374
+ 2,
1375
+ "STRING"
1376
+ ],
1377
+ [
1378
+ 35,
1379
+ 190,
1380
+ 1,
1381
+ 191,
1382
+ 0,
1383
+ "STRING"
1384
+ ]
1385
+ ],
1386
+ "groups": [
1387
+ {
1388
+ "id": 1,
1389
+ "title": "三格人物卡 / Character Card · Inputs",
1390
+ "bounding": [
1391
+ 0,
1392
+ -40,
1393
+ 1660,
1394
+ 1080
1395
+ ],
1396
+ "color": "#3f789e",
1397
+ "font_size": 24,
1398
+ "flags": {}
1399
+ },
1400
+ {
1401
+ "id": 2,
1402
+ "title": "模型与采样 / Models & Sampling",
1403
+ "bounding": [
1404
+ 0,
1405
+ 1090,
1406
+ 1750,
1407
+ 750
1408
+ ],
1409
+ "color": "#59616b",
1410
+ "font_size": 24,
1411
+ "flags": {}
1412
+ }
1413
+ ],
1414
+ "config": {},
1415
+ "extra": {
1416
+ "ds": {
1417
+ "scale": 0.8,
1418
+ "offset": [
1419
+ 90,
1420
+ 170
1421
+ ]
1422
+ },
1423
+ "character_card_candidate": {
1424
+ "release_ready": false,
1425
+ "model_math": "shared numeric spatial branch; UI packaging only",
1426
+ "initial_adapter": "character-card-v04/candidate320-single.safetensors",
1427
+ "adapter_contract": {
1428
+ "adapter_path": "/home/coffmetoer/lora_train/storyflow/character-card/v0.4-rebuild/release_candidate320/candidate320-single.safetensors",
1429
+ "package": {
1430
+ "architecture_config_sha256": "65b85015e515bbfc3ffe6c03e10537a47121bd3756e0d21a1a87126da60bc4a8",
1431
+ "inputs": {
1432
+ "/home/coffmetoer/lora_train/archive/shared/00_shared_dependencies/Krea2-Identity-Edit/krea2_identity_edit_v1_2_r64.safetensors": "f794b47142555c929cf536a2f1e4f335174b9aedbb08572b07d45814d4242423",
1433
+ "/home/coffmetoer/lora_train/storyflow/character-card/v0.4-rebuild/config/grounded_formal90_compiled.json": "268b29b1ced9a2d9c68cbe7fa0ec1e67029418ca3756f6064342f33982dd5da5",
1434
+ "/home/coffmetoer/lora_train/storyflow/character-card/v0.4-rebuild/evaluation/formal320_paired_diagnostic/r002_biker_x034_076--1024x576--s20260910--measured/request.json": "ddb04f6fb0e3a05fce3a61146a6b98c381a1f4976a029a4b570a2ba0a8d0a9c6",
1435
+ "/home/coffmetoer/lora_train/storyflow/character-card/v0.4-rebuild/workspace/grounded_formal90_compiled/save/step-000000320.safetensors": "8f93f5f445e41e82d8426f3403908be35997177712bbc1f8ff6964bdb5ffa660"
1436
+ },
1437
+ "numerical_claim": "algebraically equivalent additive deltas; finite-precision runtime is not bitwise guaranteed",
1438
+ "output_lora_dtype": "float32",
1439
+ "protocol": "identity-task-lowrank-concat-v1",
1440
+ "rank_truncation": false,
1441
+ "remove_external_identity_loader": true,
1442
+ "spatial_source_sha256": "8f93f5f445e41e82d8426f3403908be35997177712bbc1f8ff6964bdb5ffa660",
1443
+ "status": "unselected_cpu_package_not_quality_approved",
1444
+ "tool_sha256": "47056b2f47aa7a42385168c43eb40a1302a22449b4aee874fabe26002a3c91ce",
1445
+ "unit_strengths": true
1446
+ },
1447
+ "identity_loading": "bundled_identity_external71_removed"
1448
+ },
1449
+ "initial_adapter_visual_status": "candidate320 66-image baseline; portrait expansion rejected after four-panel regression",
1450
+ "global_prompt_handling": "assembled clothing/body/styling and role template",
1451
+ "region_prompt_handling": "exact original CharacterCardAppearanceInputs baseline text; UI controls only",
1452
+ "source_UI_template": "/home/coffmetoer/lora_train/storyflow/character-card/v0.3-reference-edit/release-native/character_card_workflow.json",
1453
+ "single_reference": true,
1454
+ "batch_size": 1,
1455
+ "cfg": 1,
1456
+ "fixed_three_panels": true,
1457
+ "quality_fix_claimed": false
1458
+ },
1459
+ "linearMode": false
1460
+ },
1461
+ "version": 0.4
1462
+ }
storyflow/character-card-v0.4/Storyflow-v04-support-BF16-default.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0dc8ddbaf29661e8c84c7c81019e11cfa3a0ddc390050f299ef03cb294c5869f
3
+ size 2196757
storyflow/character-card-v0.4/candidate320-single-bf16.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17d9ae50d2714b7c0d04da9a4f4f5755ac775dc28deb628a4fa51f1130d7b2ad
3
+ size 1117605928
storyflow/character-card-v0.4/candidate320-single.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2bf21aa6ddb781a5d12710e9a4930671d1d98f8f84c546d749c76f5865898647
3
+ size 2235070944
storyflow/character-card-v0.4/source.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": 2906958,
3
+ "version_id": 3314189,
4
+ "title": "Storyflow Character Card",
5
+ "source_model_name": "Storyflow Character Card",
6
+ "source_version_name": "v0.4 — Three-View Character Card",
7
+ "base_model": "Krea 2",
8
+ "type": "LORA",
9
+ "source_url": "https://civitai.red/models/2906958?modelVersionId=3314189",
10
+ "trained_words": [],
11
+ "verified_usage": [
12
+ {
13
+ "trigger": "",
14
+ "default_strength": 1
15
+ },
16
+ {
17
+ "trigger": "",
18
+ "default_strength": 1
19
+ }
20
+ ],
21
+ "source_model_description_html": "<p>September 11 — Encoder compatibility fix / 编码器兼容修复</p><p>BF16 remains the default to keep the original example settings. The encoder was not trained; I only removed the filename/precision restriction. FP8 is an optional choice, not the new default. Download the original BF16 encoder here:</p><p><a target=\"_blank\" rel=\"ugc\" href=\"https://huggingface.co/krea/Krea-2-Raw/blob/main/text_encoder/model.safetensors\">https://huggingface.co/krea/Krea-2-Raw/blob/main/text_encoder/model.safetensors</a> </p><p>Rename model.safetensors to qwen3vl_training_bf16.safetensors and place it in ComfyUI/models/text_encoders/character-card/. Hugging Face login/access may be required. The updated compatibility loader also accepts compatible standard Krea 2 encoders; select your file in its dropdown. For the optional FP8 encoder, keep its filename and put it in ComfyUI/models/text_encoders/:</p><p><a target=\"_blank\" rel=\"ugc\" href=\"https://huggingface.co/Comfy-Org/Krea-2/blob/main/text_encoders/qwen3vl_4b_fp8_scaled.safetensors\">https://huggingface.co/Comfy-Org/Krea-2/blob/main/text_encoders/qwen3vl_4b_fp8_scaled.safetensors</a> </p><p>默认仍是原来的 BF16,保留原例图设置。编码器未参与训练,这次只取消文件名和精度限制,FP8 可选,不改默认。BF16 下载第一个链接,把 model.safetensors 改名为 qwen3vl_training_bf16.safetensors,放到 ComfyUI/models/text_encoders/character-card/。可选 FP8 下载第二个链接,保留文件名放入 ComfyUI/models/text_encoders/,在编码器节点下拉框切换即可。不同精度可能改变人脸和部分布局细节。LoRA 本体没变。</p><p><span style=\"font-size:26px\"><strong>v0.4 — Three-View Character Card</strong></span></p><p>I retrained this version to keep the portrait, headless front and back views more consistent. Upload a face reference, describe the outfit and build, and pick a size. The form now has random prompt buttons too; you can edit whatever they fill in.</p><p>Use Krea 2 Turbo with the included node. Start with the <a target=\"_blank\" rel=\"ugc\" href=\"https://civitai.red/api/download/models/3314189?fileId=3201057\">support pack</a>: it includes the node, reference image, workflows and setup notes. There is also a <a target=\"_blank\" rel=\"ugc\" href=\"https://civitai.red/api/download/models/3314189?fileId=3201055\">bilingual form</a>, a <a target=\"_blank\" rel=\"ugc\" href=\"https://civitai.red/api/download/models/3314189?fileId=3201056\">regular ComfyUI JSON</a> and a <a target=\"_blank\" rel=\"ugc\" href=\"https://civitai.red/api/download/models/3314189?fileId=3200875\">node-only ZIP</a>.</p><p>The form needs ComfyUI frontend 1.41.13 or newer; the regular JSON works without App Mode. Supported sizes: 1024×576, 1280×720 and 1536×864. The gallery has a real workflow screenshot and original PNGs of a few cosplay and story characters.</p><p>A quick setup note: I combined the face-reference LoRA (Identity Edit), character-card LoRA and layout weights into one file. No need to add Identity LoRA separately. The FP32 file is about 2 GB; your reference photo is not baked in, so upload the face you want each time.</p><p>You still need Krea 2 Turbo, the text encoder, VAE, comfyui-krea2edit and the included spatial node. I wrote that spatial node for this workflow; it is not built into ComfyUI. Install it from the support pack and use the supplied workflow. A regular Load LoRA node alone will not run the whole setup. The node helps with the three-panel layout, but some prompts can still add an extra panel.</p><p>These are all beta releases for now—use whichever suits your needs. The full v1.0 release will bring character cards, expression cards and storyboard cards together with full spatial coordinate editing. It is still in training.</p><h2 id=\"v0.4\">v0.4 更新</h2><p>这版重新训了一次,主要把肖像、无头正面、背面三格稳住。传参考脸,写服装和身材,选尺寸就能跑。表单也加了随机提示词按钮,随机出来还能接着改。</p><p>用 Krea 2 Turbo,装好配套节点。先下上面的配套包,节点、参考图、工作流和安装说明都在里面;中英双语表单、普通 ComfyUI JSON 和单独节点包也能分别下载。</p><p>表单需要 ComfyUI 前端 1.41.13 或更新版,没有 App Mode 就用普通 JSON。支持 1024×576、1280×720、1536×864。展示放了一张实际界面截图,还有几个 coser、剧情人物的原生 PNG。</p><p>补充说一下,这次把人脸参考 LoRA(Identity Edit)、人物卡 LoRA 和控制布局的权重合到一个文件里了,不用再单独叠 Identity LoRA。现在存的是 FP32,所以文件有 2 GB 左右。参考照片没合进去,想用谁的脸还是自己传。</p><p>但不是下这一个 LoRA 就什么都不用装了。Krea 2 Turbo、文本编码器、VAE、comfyui-krea2edit 和配套空间节点都还要有。空间节点是我为这个工作流自己写的,不是 ComfyUI 自带的,跟着配套包装好,用给出的工作流跑就行,别只接一个普通 Load LoRA。它是帮忙稳住三格,换些提示词还是可能多出一格。</p><p>目前这些都是 beta 版本,大家按需取用。正式版 v1.0 会把现在的人物卡、表情卡和分镜卡融合到一起,支持完整的空间坐标编辑,目前还在训练中。</p><hr /><h2 id=\"earlier-versions\">Earlier versions / 旧版记录</h2><h2 id=\"v0.2-face-reference-edit\">v0.2 — Face Reference Edit</h2><p>I made this version to work from a face reference. Drop in a face, describe the outfit and build, and get a three-panel character card. I use it for my own video work; sharing it here for reference.</p><p>I put the short prompt into a small node this time, so you only need to fill in <strong>subject</strong> and <strong>presentation</strong>. The first is for age, gender and build; the second is for clothes and pose. The three-panel instructions stay inside the node.</p><p>For example, subject can be <code>a clearly adult 32-year-old woman with a healthy slim build</code>. For presentation, try <code>Wearing a plain white crew-neck T-shirt, dark blue straight jeans and white low-top sneakers, no accessories, neutral standing pose.</code></p><h3 id=\"workflow-and-files\">Workflow and files</h3><p><strong>September 9 — I repacked the node.</strong> The old ZIP bundled the node with the examples, which made the install confusing. Sorry about that. For the node alone, download <a target=\"_blank\" rel=\"ugc\" href=\"https://civitai.red/api/download/models/3298205?fileId=3193953\">the prompt node ZIP</a> and extract its folder into <code>ComfyUI/custom_nodes/</code>. Open that folder and check that <code>__init__.py</code> is directly inside:</p><pre><code>ComfyUI/custom_nodes/ComfyUI-Storyflow-Card-Prompt/__init__.py</code></pre><p>If your unzip tool adds another folder with the same name, move the inner one up a level. If you put the whole <code>Krea-Storyflow-Face-Card-v0.2</code> bundle in custom_nodes, move it out first; renaming that outer folder won't fix it. Restart ComfyUI after placing the node.</p><p>The <a target=\"_blank\" rel=\"ugc\" href=\"https://civitai.red/api/download/models/3298205?fileId=3183313\">workflow JSON</a> and <a target=\"_blank\" rel=\"ugc\" href=\"https://civitai.red/api/download/models/3298205?fileId=3193954\">full support bundle (September 9)</a> are separate. The full bundle has examples and instructions; don't install its outer folder as a node. You also need <a target=\"_blank\" rel=\"ugc\" href=\"https://github.com/lbouaraba/comfyui-krea2edit\">comfyui-krea2edit</a>. Once installed, open the workflow and edit subject and presentation as before.</p><p>The workflow now uses Moody premium. Identity Edit v1.2 uses the regular model-only loader at 1; this task LoRA uses bypass at 1. Keep the workflow's 1024×576, grounding 768, Euler, CFG 1 and fixed 8-step sigmas to start with. Model weights are separate downloads.</p><p>The new gallery pictures are the original PNGs: adult and older men and women, with both clothed and figure-study examples. The ZIP includes these originals, a reference face and setup notes. Pick a reference with the face, age and makeup you want; using other models won't replace the reference face.</p><h3 id=\"a-few-test-notes\">A few test notes</h3><p>With this short prompt, all 12 task-ON images had a headless center. In the four matched task-OFF checks, one did. Identity Edit stayed on in both. Longer prompts could also get the layout without this task LoRA, so the main benefit here is getting it with less prompt writing. The left portrait can still crop too tightly, and older bodies sometimes look younger than the face.</p><p>The earlier Turbo tests are separate: layout 31/32, identity and outfit 32/32, plus 8/8 ComfyUI bypass checks. These are just the samples I tested. The new prompt node also reproduced the matching short-prompt image pixel for pixel.</p><p>No redistribution or commercial use of the new task weights. Third-party licenses still apply. This is an unofficial modification. The face uses the \"Moody\" model,thanks to Mao-lao.</p><hr /><h2 id=\"v0.2\">v0.2 更新</h2><p>这次做了个参考脸的版本,丢一张脸进去,衣服和体型自己写,还是出三栏角色卡。也是平时自己做视频用的,仅供参考。</p><p>提示词这次顺手做成了一个小节点,只留 <strong>subject</strong> 和 <strong>presentation</strong> 两个地方要填。前一个写成年年龄、性别和体型,后一个写衣服和站姿。三栏怎么排这些固定句子已��放进去了,不用每次自己写。</p><p>上面放了可以直接填的英文例子,先照着跑,再换成自己要的衣服就行。不需要另外加触发词。</p><h3 id=\"workflow\">workflow 和配套文件</h3><p><strong>9 月 9 日,重新打了节点包。</strong> 之前把节点和示例装在一个大包里,安装时容易弄混,这里是我没处理好。现在只装节点的话,下载 <a target=\"_blank\" rel=\"ugc\" href=\"https://civitai.red/api/download/models/3298205?fileId=3193953\">提示词节点 ZIP</a>,把里面的文件夹解压到 <code>ComfyUI/custom_nodes/</code>。打开节点文件夹,应该直接看到 <code>__init__.py</code>:</p><pre><code>ComfyUI/custom_nodes/ComfyUI-Storyflow-Card-Prompt/__init__.py</code></pre><p>有的解压软件会多套一层同名目录,遇到这种情况把内层挪出来。如果之前把整个 <code>Krea-Storyflow-Face-Card-v0.2</code> 放进了 custom_nodes,先把它移出去;只改外层文件夹的名字没有用。放好真正的节点文件夹后再重启。</p><p><a target=\"_blank\" rel=\"ugc\" href=\"https://civitai.red/api/download/models/3298205?fileId=3183313\">workflow JSON</a> 和<a target=\"_blank\" rel=\"ugc\" href=\"https://civitai.red/api/download/models/3298205?fileId=3193954\">完整配套包(9 月 9 日)</a>单独下载。完整配套包里是示例和说明,不能把它的外层当节点安装。另外仍然需要 <a target=\"_blank\" rel=\"ugc\" href=\"https://github.com/lbouaraba/comfyui-krea2edit\">comfyui-krea2edit</a>。装好以后照常打开 workflow,改 subject 和 presentation 就行。</p><p>workflow 现在默认用 Moody premium。Identity Edit v1.2 普通加载,强度 1;这次的任务 LoRA 用 bypass,强度也是 1。1024×576、grounding 768、Euler、CFG 1 和固定 8 步 sigmas 先按 workflow 来。模型权重另外下载,ZIP 里没有放。</p><p>新加的都是原始 PNG,成年男女、老年男女都跑了穿衣和人体艺术图。原文件、参考脸和安装说明也放进 ZIP 了。想要什么脸、年龄和妆容,先选对应参考脸;换底模不会自动把参考脸也换掉。</p><h3>简单说下实测</h3><p>用这个短提示词,开着任务 LoRA 跑的 12 张中间都没长脸;同条件关掉任务 LoRA 的四组对照里,有一组也能做到。两边都保留了 Identity Edit。写长一点的提示词,底模本身也能画出来,所以这个 LoRA 主要是让短输入更容易保持三栏结构。左边肖像有时裁得太近,老年人的身体也偶尔显年轻,这些还不完美。</p><p>之前 Turbo 那轮单独记:格式 31/32,身份和衣服 32/32,另外 ComfyUI bypass 测了 8 张,8 张通过。这里只是自己测过的样本。新节点也实际跑过,和直接填短提示词的那张图逐像素一样。</p><p>新增任务权重禁止二次分发和商用,第三方原有许可照旧。这是非官方修改成果,人脸使用moody模型,感谢猫佬。</p><hr /><p></p><p><span style=\"color:rgb(134, 142, 150)\">Older version — just for reference</span></p><p><span style=\"color:rgb(134, 142, 150)\">The earlier version is a text-to-image LoKr for three-panel cards, tested on Krea 2 Turbo BF16 and Moody-Krea BF16. Use a regular Load LoRA node at 0.7, 1536 × 864, 8 steps, CFG 1, Euler / Beta. Its trigger is single_face_character_card. For v0.2, use the workflow above.</span></p><p><span style=\"color:rgb(134, 142, 150)\">旧版说明(留个记录)</span></p><p><span style=\"color:rgb(134, 142, 150)\">以前那版是纯文生图,用 LoKr 做三栏角色卡,在 Krea 2 Turbo BF16 和 Moody-Krea BF16 上试过。普通 Load LoRA 节点加载,强度 0.7,1536 × 864、8 steps、CFG 1、Euler / Beta。</span></p><p><span style=\"color:rgb(134, 142, 150)\">旧版触发词是 single_face_character_card,只给以前那版用。现在用 v0.2 的话,直接看上面的 workflow 就好。</span></p>",
22
+ "source_version_description_html": "<p>Added a BF16 precision variant of this LoRA: 1.04 GB, half the FP32 size. Three same-reference, prompt and seed comparisons at 1024×576, 1280×720 and 1536×864 matched the FP32 output pixel for pixel. Select BF16 in this version’s download selector. Put candidate320-single-bf16.safetensors in ComfyUI/models/loras/character-card-v04/ and choose it in the spatial node’s adapter_name dropdown. The existing workflows work with both files; FP32 and the text encoder default stay unchanged.</p><p>补了同版本的 BF16 LoRA,约 1.04 GB,体积减半。三个尺寸同参考、同描述、同种子对比,三组都和 FP32 逐像素一致。下载处直接选 BF16,文件放进 ComfyUI/models/loras/character-card-v04/,在空间节点的 adapter_name 下拉框选 candidate320-single-bf16.safetensors 就行。现有工作流两种都能用,FP32 继续保留,文本编码器默认不变。</p><p>September 11 fix: BF16 remains the default, preserving the original example settings. The text encoder was not trained; I removed the unnecessary filename/precision restriction. FP8 is optional. Download the original BF16 encoder below, rename model.safetensors to qwen3vl_training_bf16.safetensors, and put it in ComfyUI/models/text_encoders/character-card/.</p><p><a target=\"_blank\" rel=\"ugc\" href=\"https://huggingface.co/krea/Krea-2-Raw/blob/main/text_encoder/model.safetensors\">https://huggingface.co/krea/Krea-2-Raw/blob/main/text_encoder/model.safetensors</a></p><p>9 月 11 日修复:默认仍用原来的 BF16,保留原例图设置。编码器没参与训练,这次只取消文件名和精度限制,FP8 可以自己选。BF16 下载上面的文件,改名为 qwen3vl_training_bf16.safetensors,放进 ComfyUI/models/text_encoders/character-card/。重新下载带 BF16-default 的工作流和配套包即可,LoRA 本体没变。可选 FP8 的下载链接也补在主页说明里了。</p><p>I retrained this version to keep the portrait, headless front and back views more consistent. Upload a face, write the outfit and build, and pick a size. I also added a bilingual form, random prompt buttons and a regular ComfyUI workflow, with a few cosplay and story-character examples.</p><p>A quick setup note: I put the Identity LoRA, character-card LoRA and layout weights into one file, so you don't need to add Identity LoRA separately. That's also why this FP32 file is about 2 GB. Your reference photo isn't baked in—upload the face you want each time.</p><p>You still need Krea 2 Turbo, the text encoder, VAE, comfyui-krea2edit and the included spatial node. I wrote that spatial node for this workflow; it isn't built into ComfyUI. Use the supplied workflow, rather than just dropping the file into a regular Load LoRA node. It helps keep the three panels in place, but some prompts can still produce an extra panel.</p><p>Start with the support ZIP; model links and installation steps are inside. The form needs ComfyUI frontend 1.41.13 or newer; otherwise use the regular JSON. Supported sizes: 1024×576, 1280×720 and 1536×864.</p><p>这版重新训了一次,主要把肖像、无头正面、背面三格稳住。传参考脸,写服装和身材,选尺寸就能跑。中英双语表单、随机提示词按钮和普通 ComfyUI 工作流都放了,也做了几张 coser 和剧情人物的例图。</p><p>补充说一下,这次把人脸 Identity LoRA、人物卡 LoRA 和控制布局的权重合到一个文件里了,不用再单独叠 Identity LoRA。现在存的是 FP32,所以文件有 2 GB 左右。参考照片没合进去,想用谁的脸还是自己传。</p><p>但不是下这一个 LoRA 就什么都不用装了。Krea 2 Turbo、文本编码器、VAE、comfyui-krea2edit 和配套空间节点都还要有。空间节点是我为这个工作流自己写的,不是 ComfyUI 自带的,跟着配套包装好,用给出的工作流跑就行,别只接一个普通 Load LoRA。它是帮忙稳住三格,换些提示词还是可能多出一格。</p><p>先下配套 ZIP,模型链接和安装步骤都在里面。表单需要 ComfyUI 前端 1.41.13 或更新版,没有表单模式就用普通 JSON。支持 1024×576、1280×720、1536×864。</p><p>These are all beta releases for now—use whichever suits your needs. The full v1.0 release will bring character cards, expression cards and storyboard cards together with full spatial coordinate editing. It is still in training.</p><p>目前这些都是 beta 版本,大家按需取用。正式版 v1.0 会把现在的人物卡、表情卡和分镜卡融合到一起,支持完整的空间坐标编辑,目前还在训练中。</p>",
23
+ "permissions": {
24
+ "allowNoCredit": true,
25
+ "allowCommercialUse": [
26
+ "Image",
27
+ "RentCivit",
28
+ "Rent",
29
+ "Sell",
30
+ "SellMerge"
31
+ ],
32
+ "allowDerivatives": true,
33
+ "allowDifferentLicense": true
34
+ },
35
+ "files": [
36
+ {
37
+ "id": 3200875,
38
+ "name": "Storyflow-v04-spatial-node-encoder-fix.zip",
39
+ "type": "Other",
40
+ "path": "storyflow/character-card-v0.4/Storyflow-v04-spatial-node-encoder-fix.zip",
41
+ "sha256": "7e4ad3f584886e871015014a4156d01633578fe12df974fa170ac15407171604",
42
+ "size_kb": 24.9228515625,
43
+ "metadata": {
44
+ "format": "Other",
45
+ "isRequired": false
46
+ },
47
+ "download_url": "https://civitai.red/api/download/models/3314189?fileId=3200875"
48
+ },
49
+ {
50
+ "id": 3201057,
51
+ "name": "Storyflow-v04-support-BF16-default.zip",
52
+ "type": "Other",
53
+ "path": "storyflow/character-card-v0.4/Storyflow-v04-support-BF16-default.zip",
54
+ "sha256": "0dc8ddbaf29661e8c84c7c81019e11cfa3a0ddc390050f299ef03cb294c5869f",
55
+ "size_kb": 2145.2705078125,
56
+ "metadata": {
57
+ "format": "Other",
58
+ "isRequired": false
59
+ },
60
+ "download_url": "https://civitai.red/api/download/models/3314189?fileId=3201057"
61
+ },
62
+ {
63
+ "id": 3201055,
64
+ "name": "Storyflow-v04-form-BF16-default.json",
65
+ "type": "Config",
66
+ "path": "storyflow/character-card-v0.4/Storyflow-v04-form-BF16-default.json",
67
+ "sha256": "89e3edfb8c75b2fbb0a1bfd7adb28549baf60c77acddaedebd06564e548d0d65",
68
+ "size_kb": 29.6796875,
69
+ "metadata": {
70
+ "format": "Other",
71
+ "isRequired": false
72
+ },
73
+ "download_url": "https://civitai.red/api/download/models/3314189?fileId=3201055"
74
+ },
75
+ {
76
+ "id": 3201056,
77
+ "name": "Storyflow-v04-standard-BF16-default.json",
78
+ "type": "Config",
79
+ "path": "storyflow/character-card-v0.4/Storyflow-v04-standard-BF16-default.json",
80
+ "sha256": "7e31c93e8b3256633861b3e9e4fa934ca4e8b7a2b168ca76f90e45dd549b0134",
81
+ "size_kb": 27.9453125,
82
+ "metadata": {
83
+ "format": "Other",
84
+ "isRequired": false
85
+ },
86
+ "download_url": "https://civitai.red/api/download/models/3314189?fileId=3201056"
87
+ },
88
+ {
89
+ "id": 3201146,
90
+ "name": "candidate320-single-bf16.safetensors",
91
+ "type": "Model",
92
+ "path": "storyflow/character-card-v0.4/candidate320-single-bf16.safetensors",
93
+ "sha256": "17d9ae50d2714b7c0d04da9a4f4f5755ac775dc28deb628a4fa51f1130d7b2ad",
94
+ "size_kb": 1091412.0390625,
95
+ "metadata": {
96
+ "format": "SafeTensor",
97
+ "fp": "bf16",
98
+ "isRequired": false
99
+ },
100
+ "download_url": "https://civitai.red/api/download/models/3314189?fileId=3201146"
101
+ },
102
+ {
103
+ "id": 3199663,
104
+ "name": "candidate320-single.safetensors",
105
+ "type": "Model",
106
+ "path": "storyflow/character-card-v0.4/candidate320-single.safetensors",
107
+ "sha256": "2bf21aa6ddb781a5d12710e9a4930671d1d98f8f84c546d749c76f5865898647",
108
+ "size_kb": 2182686.46875,
109
+ "metadata": {
110
+ "format": "SafeTensor",
111
+ "fp": "fp32",
112
+ "isRequired": false
113
+ },
114
+ "download_url": "https://civitai.red/api/download/models/3314189?fileId=3199663"
115
+ }
116
+ ]
117
+ }