-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathgradio.log
More file actions
3593 lines (3443 loc) · 499 KB
/
gradio.log
File metadata and controls
3593 lines (3443 loc) · 499 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
2025-11-12 17:32:23.172 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 17:32:24.592 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: church, full view, negative: low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background, eval:
2025-11-12 17:32:24.592 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 17:32:24.592 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 17:37:00.387 | INFO | __main__:call_choose_model:294 - [Gradio] Choosing model for prompt: water bottle
2025-11-12 17:37:00.726 | INFO | __main__:call_choose_model:299 - [Gradio] LLM chose model: sd
2025-11-12 17:37:00.944 | INFO | __main__:_reset_ui_and_set_steps:308 - [Gradio] Resetting UI, setting steps to 28
2025-11-12 17:37:01.247 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: water bottle
2025-11-12 17:37:01.248 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: water bottle (using qwen2.5:7b)
2025-11-12 17:37:02.005 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'water bottle, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-12 17:37:02.217 | INFO | __main__:route_generation:328 - [Gradio] Routing generation to model: sd with 28 steps
2025-11-12 17:37:02.218 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 28 inference steps...
2025-11-12 17:37:02.218 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 28 steps...
2025-11-12 17:37:14.078 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-12 17:37:14.136 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_mv7ic9nh for Trellis step
2025-11-12 17:37:14.136 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-12 17:37:14.560 | INFO | __main__:route_post_processing:337 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-12 17:37:14.561 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: water bottle, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-12 17:37:14.564 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-12 17:37:14.568 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 17:37:14.571 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 17:37:19.133 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'The images match the prompt well, showing a water bottle from different angles. The images are consistent across the views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-12 17:37:19.134 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'The images match the prompt well, showing a water bottle from different angles. The images are consistent across the views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-12 17:37:19.134 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: water bottle, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval: The images match the prompt well, showing a water bottle from different angles. The images are consistent across the views, and there are no visible issues or artifacts.
2025-11-12 17:37:19.134 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 17:37:19.134 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 17:37:25.867 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: water bottle, full view, negative: low detail, bad proportions, deformed parts, artifacts, shadows, lighting, background, eval:
2025-11-12 17:37:25.867 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 17:37:25.867 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 17:37:28.254 | INFO | __main__:call_choose_model:294 - [Gradio] Choosing model for prompt: green water bottle
2025-11-12 17:37:28.530 | INFO | __main__:call_choose_model:299 - [Gradio] LLM chose model: sd
2025-11-12 17:37:28.759 | INFO | __main__:_reset_ui_and_set_steps:308 - [Gradio] Resetting UI, setting steps to 28
2025-11-12 17:37:29.062 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: green water bottle
2025-11-12 17:37:29.062 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: green water bottle (using qwen2.5:7b)
2025-11-12 17:37:29.789 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'water bottle, green, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-12 17:37:29.974 | INFO | __main__:route_generation:328 - [Gradio] Routing generation to model: sd with 28 steps
2025-11-12 17:37:29.975 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 28 inference steps...
2025-11-12 17:37:29.975 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 28 steps...
2025-11-12 17:37:41.841 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-12 17:37:41.890 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_1k9lef1b for Trellis step
2025-11-12 17:37:41.891 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-12 17:37:42.305 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: water bottle, green, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-12 17:37:42.306 | INFO | __main__:route_post_processing:337 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-12 17:37:42.308 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 17:37:42.310 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-12 17:37:42.311 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 17:37:46.731 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'the images match the prompt well. The images are consistent across the multi-views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-12 17:37:46.732 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'the images match the prompt well. The images are consistent across the multi-views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-12 17:37:46.732 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: water bottle, green, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval: the images match the prompt well. The images are consistent across the multi-views, and there are no visible issues or artifacts.
2025-11-12 17:37:46.732 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 17:37:46.732 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 17:37:53.519 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: water bottle, green, negative: bad proportions, deformed parts, shadows, blocky water surface, unrealistic reflections, over-saturated colors, background, eval:
2025-11-12 17:37:53.520 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 17:37:53.520 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 17:39:34.285 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/views_1k9lef1b/trellis_output.zip
2025-11-12 17:39:34.350 | INFO | __main__:generate_3d:246 - [TRELLIS/API] Extracted ZIP to /tmp/views_1k9lef1b
2025-11-12 17:39:34.350 | INFO | __main__:generate_3d:249 - [TRELLIS/API-DEBUG] Files in /tmp/views_1k9lef1b: ['view_00.jpg', 'view_01.jpg', 'view_02.jpg', 'view_03.jpg', 'view_04.jpg', 'view_05.jpg', 'trellis_output.zip', 'mesh_9999a711-fd76-42be-816b-d3cf63e16962.glb', 'mesh_9999a711-fd76-42be-816b-d3cf63e16962.ply', 'preview_9999a711-fd76-42be-816b-d3cf63e16962.mp4', 'frame_9999a711-fd76-42be-816b-d3cf63e16962.jpg']
2025-11-12 17:39:34.350 | INFO | __main__:generate_3d:256 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_9999a711-fd76-42be-816b-d3cf63e16962.glb']
2025-11-12 17:39:34.350 | INFO | __main__:generate_3d:257 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_9999a711-fd76-42be-816b-d3cf63e16962.ply']
2025-11-12 17:39:34.350 | INFO | __main__:generate_3d:258 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_9999a711-fd76-42be-816b-d3cf63e16962.mp4']
2025-11-12 17:39:34.350 | INFO | __main__:generate_3d:259 - [TRELLIS/API-DEBUG] Found frame files: ['frame_9999a711-fd76-42be-816b-d3cf63e16962.jpg']
2025-11-12 17:39:34.350 | INFO | __main__:generate_3d:273 - [TRELLIS/API] Frame path found: /tmp/views_1k9lef1b/frame_9999a711-fd76-42be-816b-d3cf63e16962.jpg
2025-11-12 17:39:34.350 | INFO | __main__:generate_3d:277 - [TRELLIS/API] Assets found: GLB=/tmp/views_1k9lef1b/mesh_9999a711-fd76-42be-816b-d3cf63e16962.glb, PLY=/tmp/views_1k9lef1b/mesh_9999a711-fd76-42be-816b-d3cf63e16962.ply, MP4=/tmp/views_1k9lef1b/preview_9999a711-fd76-42be-816b-d3cf63e16962.mp4
2025-11-12 18:22:52.744 | INFO | __main__:call_choose_model:294 - [Gradio] Choosing model for prompt: house
2025-11-12 18:22:55.626 | INFO | __main__:call_choose_model:299 - [Gradio] LLM chose model: flux
2025-11-12 18:22:55.836 | INFO | __main__:_reset_ui_and_set_steps:308 - [Gradio] Resetting UI, setting steps to 10
2025-11-12 18:22:56.139 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: house
2025-11-12 18:22:56.139 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: house (using qwen2.5:7b)
2025-11-12 18:22:56.935 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'house, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-12 18:22:57.155 | INFO | __main__:route_generation:328 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-12 18:22:57.155 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-12 18:22:57.155 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-12 18:23:02.389 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-12 18:23:02.402 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_d0_a7bmx for feedback
2025-11-12 18:23:02.402 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-12 18:23:02.831 | INFO | __main__:route_post_processing:353 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-12 18:23:02.832 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-12 18:23:02.834 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 18:23:02.839 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 18:23:04.155 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, full view, negative: bad proportions, deformed parts, floating objects, unnatural shadows, overcast lighting, cluttered background, blocky textures, artifacts, eval:
2025-11-12 18:23:04.155 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 18:23:04.156 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 18:24:06.815 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/flux_view_d0_a7bmx/trellis_output.zip
2025-11-12 18:24:06.877 | INFO | __main__:generate_3d:246 - [TRELLIS/API] Extracted ZIP to /tmp/flux_view_d0_a7bmx
2025-11-12 18:24:06.877 | INFO | __main__:generate_3d:249 - [TRELLIS/API-DEBUG] Files in /tmp/flux_view_d0_a7bmx: ['view_00.jpg', 'trellis_output.zip', 'mesh_26806813-72bf-4cf8-8dac-42450bb1f6cf.glb', 'mesh_26806813-72bf-4cf8-8dac-42450bb1f6cf.ply', 'preview_26806813-72bf-4cf8-8dac-42450bb1f6cf.mp4', 'frame_26806813-72bf-4cf8-8dac-42450bb1f6cf.jpg']
2025-11-12 18:24:06.877 | INFO | __main__:generate_3d:256 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_26806813-72bf-4cf8-8dac-42450bb1f6cf.glb']
2025-11-12 18:24:06.877 | INFO | __main__:generate_3d:257 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_26806813-72bf-4cf8-8dac-42450bb1f6cf.ply']
2025-11-12 18:24:06.877 | INFO | __main__:generate_3d:258 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_26806813-72bf-4cf8-8dac-42450bb1f6cf.mp4']
2025-11-12 18:24:06.877 | INFO | __main__:generate_3d:259 - [TRELLIS/API-DEBUG] Found frame files: ['frame_26806813-72bf-4cf8-8dac-42450bb1f6cf.jpg']
2025-11-12 18:24:06.877 | INFO | __main__:generate_3d:273 - [TRELLIS/API] Frame path found: /tmp/flux_view_d0_a7bmx/frame_26806813-72bf-4cf8-8dac-42450bb1f6cf.jpg
2025-11-12 18:24:06.877 | INFO | __main__:generate_3d:277 - [TRELLIS/API] Assets found: GLB=/tmp/flux_view_d0_a7bmx/mesh_26806813-72bf-4cf8-8dac-42450bb1f6cf.glb, PLY=/tmp/flux_view_d0_a7bmx/mesh_26806813-72bf-4cf8-8dac-42450bb1f6cf.ply, MP4=/tmp/flux_view_d0_a7bmx/preview_26806813-72bf-4cf8-8dac-42450bb1f6cf.mp4
2025-11-12 19:23:42.676 | INFO | __main__:call_choose_model:294 - [Gradio] Choosing model for prompt: house
2025-11-12 19:23:42.960 | INFO | __main__:call_choose_model:299 - [Gradio] LLM chose model: flux
2025-11-12 19:23:43.182 | INFO | __main__:_reset_ui_and_set_steps:308 - [Gradio] Resetting UI, setting steps to 10
2025-11-12 19:23:43.483 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: house
2025-11-12 19:23:43.484 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: house (using qwen2.5:7b)
2025-11-12 19:23:44.208 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'house, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-12 19:23:44.394 | INFO | __main__:route_generation:328 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-12 19:23:44.394 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-12 19:23:44.394 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-12 19:23:49.632 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-12 19:23:49.643 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_1zau3y9a for feedback
2025-11-12 19:23:49.643 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-12 19:23:55.803 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-12 19:23:55.804 | INFO | __main__:route_post_processing:353 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-12 19:23:55.804 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 19:23:55.809 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 19:23:57.119 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, full view, negative: bad proportions, artifacts, shadows, lighting, background, over-saturated colors, repetitive textures, eval:
2025-11-12 19:23:57.119 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 19:23:57.119 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 19:34:00.341 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/flux_view_1zau3y9a/trellis_output.zip
2025-11-12 19:34:00.455 | INFO | __main__:generate_3d:246 - [TRELLIS/API] Extracted ZIP to /tmp/flux_view_1zau3y9a
2025-11-12 19:34:00.455 | INFO | __main__:generate_3d:249 - [TRELLIS/API-DEBUG] Files in /tmp/flux_view_1zau3y9a: ['view_00.jpg', 'trellis_output.zip', 'mesh_85cafcc0-48e4-4263-bcb5-86d5178ba366.glb', 'mesh_85cafcc0-48e4-4263-bcb5-86d5178ba366.ply', 'preview_85cafcc0-48e4-4263-bcb5-86d5178ba366.mp4', 'frame_85cafcc0-48e4-4263-bcb5-86d5178ba366.jpg']
2025-11-12 19:34:00.455 | INFO | __main__:generate_3d:256 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_85cafcc0-48e4-4263-bcb5-86d5178ba366.glb']
2025-11-12 19:34:00.455 | INFO | __main__:generate_3d:257 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_85cafcc0-48e4-4263-bcb5-86d5178ba366.ply']
2025-11-12 19:34:00.456 | INFO | __main__:generate_3d:258 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_85cafcc0-48e4-4263-bcb5-86d5178ba366.mp4']
2025-11-12 19:34:00.456 | INFO | __main__:generate_3d:259 - [TRELLIS/API-DEBUG] Found frame files: ['frame_85cafcc0-48e4-4263-bcb5-86d5178ba366.jpg']
2025-11-12 19:34:00.456 | INFO | __main__:generate_3d:273 - [TRELLIS/API] Frame path found: /tmp/flux_view_1zau3y9a/frame_85cafcc0-48e4-4263-bcb5-86d5178ba366.jpg
2025-11-12 19:34:00.456 | INFO | __main__:generate_3d:277 - [TRELLIS/API] Assets found: GLB=/tmp/flux_view_1zau3y9a/mesh_85cafcc0-48e4-4263-bcb5-86d5178ba366.glb, PLY=/tmp/flux_view_1zau3y9a/mesh_85cafcc0-48e4-4263-bcb5-86d5178ba366.ply, MP4=/tmp/flux_view_1zau3y9a/preview_85cafcc0-48e4-4263-bcb5-86d5178ba366.mp4
2025-11-12 19:40:22.216 | INFO | __main__:call_choose_model:294 - [Gradio] Choosing model for prompt: green water bottle
2025-11-12 19:40:24.983 | INFO | __main__:call_choose_model:299 - [Gradio] LLM chose model: sd
2025-11-12 19:40:25.206 | INFO | __main__:_reset_ui_and_set_steps:308 - [Gradio] Resetting UI, setting steps to 28
2025-11-12 19:40:25.561 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: green water bottle
2025-11-12 19:40:25.561 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: green water bottle (using qwen2.5:7b)
2025-11-12 19:40:26.313 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'water bottle, green, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-12 19:40:26.524 | INFO | __main__:route_generation:328 - [Gradio] Routing generation to model: sd with 28 steps
2025-11-12 19:40:26.524 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 28 inference steps...
2025-11-12 19:40:26.524 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 28 steps...
2025-11-12 19:40:38.428 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-12 19:40:38.455 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_q5vet2u7 for Trellis step
2025-11-12 19:40:38.455 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-12 19:51:36.295 | INFO | __main__:route_post_processing:337 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-12 19:51:36.296 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: water bottle, green, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-12 19:51:36.303 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 19:51:36.303 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 19:51:36.306 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-12 19:51:37.812 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: water bottle, green, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background, eval:
2025-11-12 19:51:37.812 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 19:51:37.812 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 19:51:42.174 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'the images match the prompt well. The images are consistent across the multi-views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-12 19:51:42.174 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'the images match the prompt well. The images are consistent across the multi-views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-12 19:51:42.174 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: water bottle, green, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval: the images match the prompt well. The images are consistent across the multi-views, and there are no visible issues or artifacts.
2025-11-12 19:51:42.174 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-12 19:51:42.174 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-12 19:52:19.193 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/views_q5vet2u7/trellis_output.zip
2025-11-12 19:52:19.226 | INFO | __main__:generate_3d:246 - [TRELLIS/API] Extracted ZIP to /tmp/views_q5vet2u7
2025-11-12 19:52:19.226 | INFO | __main__:generate_3d:249 - [TRELLIS/API-DEBUG] Files in /tmp/views_q5vet2u7: ['view_00.jpg', 'view_01.jpg', 'view_02.jpg', 'view_03.jpg', 'view_04.jpg', 'view_05.jpg', 'trellis_output.zip', 'mesh_94d37d13-acce-4f9f-bebf-b5d3f2d6792a.glb', 'mesh_94d37d13-acce-4f9f-bebf-b5d3f2d6792a.ply', 'preview_94d37d13-acce-4f9f-bebf-b5d3f2d6792a.mp4', 'frame_94d37d13-acce-4f9f-bebf-b5d3f2d6792a.jpg']
2025-11-12 19:52:19.227 | INFO | __main__:generate_3d:256 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_94d37d13-acce-4f9f-bebf-b5d3f2d6792a.glb']
2025-11-12 19:52:19.227 | INFO | __main__:generate_3d:257 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_94d37d13-acce-4f9f-bebf-b5d3f2d6792a.ply']
2025-11-12 19:52:19.227 | INFO | __main__:generate_3d:258 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_94d37d13-acce-4f9f-bebf-b5d3f2d6792a.mp4']
2025-11-12 19:52:19.227 | INFO | __main__:generate_3d:259 - [TRELLIS/API-DEBUG] Found frame files: ['frame_94d37d13-acce-4f9f-bebf-b5d3f2d6792a.jpg']
2025-11-12 19:52:19.227 | INFO | __main__:generate_3d:273 - [TRELLIS/API] Frame path found: /tmp/views_q5vet2u7/frame_94d37d13-acce-4f9f-bebf-b5d3f2d6792a.jpg
2025-11-12 19:52:19.227 | INFO | __main__:generate_3d:277 - [TRELLIS/API] Assets found: GLB=/tmp/views_q5vet2u7/mesh_94d37d13-acce-4f9f-bebf-b5d3f2d6792a.glb, PLY=/tmp/views_q5vet2u7/mesh_94d37d13-acce-4f9f-bebf-b5d3f2d6792a.ply, MP4=/tmp/views_q5vet2u7/preview_94d37d13-acce-4f9f-bebf-b5d3f2d6792a.mp4
2025-11-13 11:10:30.546 | INFO | __main__:call_choose_model:294 - [Gradio] Choosing model for prompt: A small square cabin
2025-11-13 11:10:33.102 | INFO | __main__:call_choose_model:299 - [Gradio] LLM chose model: flux
2025-11-13 11:10:33.334 | INFO | __main__:_reset_ui_and_set_steps:308 - [Gradio] Resetting UI, setting steps to 10
2025-11-13 11:10:33.487 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: A small square cabin
2025-11-13 11:10:33.487 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: A small square cabin (using qwen2.5:7b)
2025-11-13 11:10:34.310 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'cabin, small, square, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-13 11:10:34.451 | INFO | __main__:route_generation:328 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 11:10:34.451 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 11:10:34.451 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 11:10:39.403 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 11:10:39.410 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_wenx1us4 for feedback
2025-11-13 11:10:39.410 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 11:10:39.624 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: cabin, small, square, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-13 11:10:39.624 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:10:39.626 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:10:39.674 | INFO | __main__:route_post_processing:353 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 11:10:41.046 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: cabin, small, square, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, unrealistic lighting, cluttered background, eval:
2025-11-13 11:10:41.046 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:10:41.046 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:10:44.535 | INFO | __main__:route_generation:328 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 11:10:44.535 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 11:10:44.535 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 11:10:49.491 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 11:10:49.498 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_hfy4l7j9 for feedback
2025-11-13 11:10:49.498 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 11:10:49.817 | INFO | __main__:route_post_processing:353 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 11:10:49.817 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: cabin, small, square, full view, negative: low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, unrealistic lighting, cluttered background, eval:
2025-11-13 11:10:49.818 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:10:49.819 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:10:50.881 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: cabin, small, square, full view, negative: low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, unrealistic lighting, cluttered background, eval:
2025-11-13 11:10:50.882 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:10:50.882 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:10:53.872 | INFO | __main__:route_generation:328 - [Gradio] Routing generation to model: flux with 15 steps
2025-11-13 11:10:53.872 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 15 inference steps...
2025-11-13 11:10:53.873 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 15 steps...
2025-11-13 11:11:01.093 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 11:11:01.099 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_q7k8pon3 for feedback
2025-11-13 11:11:01.099 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 11:11:01.410 | INFO | __main__:route_post_processing:353 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 11:11:01.410 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: cabin, small, square, full view, negative: low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, unrealistic lighting, cluttered background, eval:
2025-11-13 11:11:01.411 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:11:01.411 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:11:02.421 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: cabin, small, square, full view, negative: low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, unrealistic lighting, cluttered background, eval:
2025-11-13 11:11:02.421 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:11:02.421 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:12:06.034 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/flux_view_q7k8pon3/trellis_output.zip
2025-11-13 11:12:06.103 | INFO | __main__:generate_3d:246 - [TRELLIS/API] Extracted ZIP to /tmp/flux_view_q7k8pon3
2025-11-13 11:12:06.104 | INFO | __main__:generate_3d:249 - [TRELLIS/API-DEBUG] Files in /tmp/flux_view_q7k8pon3: ['view_00.jpg', 'trellis_output.zip', 'mesh_25d58e6a-a0c0-4b42-ad32-f6d533da58cd.glb', 'mesh_25d58e6a-a0c0-4b42-ad32-f6d533da58cd.ply', 'preview_25d58e6a-a0c0-4b42-ad32-f6d533da58cd.mp4', 'frame_25d58e6a-a0c0-4b42-ad32-f6d533da58cd.jpg']
2025-11-13 11:12:06.104 | INFO | __main__:generate_3d:256 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_25d58e6a-a0c0-4b42-ad32-f6d533da58cd.glb']
2025-11-13 11:12:06.104 | INFO | __main__:generate_3d:257 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_25d58e6a-a0c0-4b42-ad32-f6d533da58cd.ply']
2025-11-13 11:12:06.104 | INFO | __main__:generate_3d:258 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_25d58e6a-a0c0-4b42-ad32-f6d533da58cd.mp4']
2025-11-13 11:12:06.104 | INFO | __main__:generate_3d:259 - [TRELLIS/API-DEBUG] Found frame files: ['frame_25d58e6a-a0c0-4b42-ad32-f6d533da58cd.jpg']
2025-11-13 11:12:06.104 | INFO | __main__:generate_3d:273 - [TRELLIS/API] Frame path found: /tmp/flux_view_q7k8pon3/frame_25d58e6a-a0c0-4b42-ad32-f6d533da58cd.jpg
2025-11-13 11:12:06.104 | INFO | __main__:generate_3d:277 - [TRELLIS/API] Assets found: GLB=/tmp/flux_view_q7k8pon3/mesh_25d58e6a-a0c0-4b42-ad32-f6d533da58cd.glb, PLY=/tmp/flux_view_q7k8pon3/mesh_25d58e6a-a0c0-4b42-ad32-f6d533da58cd.ply, MP4=/tmp/flux_view_q7k8pon3/preview_25d58e6a-a0c0-4b42-ad32-f6d533da58cd.mp4
2025-11-13 11:12:14.372 | INFO | __main__:send_review:473 - [FEEDBACK] Success: Feedback sent!
2025-11-13 11:13:20.063 | INFO | __main__:call_choose_model:294 - [Gradio] Choosing model for prompt: A large rabbit with red hair holding a bazooka
2025-11-13 11:13:20.401 | INFO | __main__:call_choose_model:299 - [Gradio] LLM chose model: sd
2025-11-13 11:13:20.619 | INFO | __main__:_reset_ui_and_set_steps:308 - [Gradio] Resetting UI, setting steps to 28
2025-11-13 11:13:20.773 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: A large rabbit with red hair holding a bazooka
2025-11-13 11:13:20.773 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: A large rabbit with red hair holding a bazooka (using qwen2.5:7b)
2025-11-13 11:13:21.638 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'rabbit, large, with red hair, holding a bazooka, full body', 'negative': 'blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background'}
2025-11-13 11:13:21.787 | INFO | __main__:route_generation:328 - [Gradio] Routing generation to model: sd with 28 steps
2025-11-13 11:13:21.787 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 28 inference steps...
2025-11-13 11:13:21.788 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 28 steps...
2025-11-13 11:13:33.664 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 11:13:33.717 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_xw6x2__7 for Trellis step
2025-11-13 11:13:33.717 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 11:13:34.040 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: rabbit, large, with red hair, holding a bazooka, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval:
2025-11-13 11:13:34.042 | INFO | __main__:route_post_processing:337 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-13 11:13:34.044 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:13:34.049 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:13:34.051 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-13 11:13:35.372 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: rabbit, large, with red hair, holding a bazooka, full body, negative: low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, unnatural shadows, unrealistic lighting, background, eval:
2025-11-13 11:13:35.374 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:13:35.374 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:13:39.671 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'The images match the prompt well, showing a large rabbit with red hair holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 11:13:39.672 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'The images match the prompt well, showing a large rabbit with red hair holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 11:13:39.672 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: rabbit, large, with red hair, holding a bazooka, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval: The images match the prompt well, showing a large rabbit with red hair holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.
2025-11-13 11:13:39.672 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:13:39.672 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:13:49.180 | INFO | __main__:route_generation:328 - [Gradio] Routing generation to model: sd with 28 steps
2025-11-13 11:13:49.181 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 28 inference steps...
2025-11-13 11:13:49.181 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 28 steps...
2025-11-13 11:14:01.089 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 11:14:01.141 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_ft8d9pkq for Trellis step
2025-11-13 11:14:01.142 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 11:14:01.460 | INFO | __main__:route_post_processing:337 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-13 11:14:01.461 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: rabbit, large, with red hair, holding a bazooka, full body, negative: blurry, low detail, bad anatomy, extra limbs, deformed face, deformed tail, repetitive foliage, tiling textures, lighting, shadows, background, eval:
2025-11-13 11:14:01.465 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:14:01.466 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-13 11:14:01.468 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:14:02.829 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: rabbit, large, with red hair, holding a bazooka, full body, negative: deformed tail, repetitive foliage, tiling textures, unnatural colors, over-saturated colors, flat water, blocky water surface, unrealistic reflections, lighting, shadows, background, eval:
2025-11-13 11:14:02.829 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:14:02.829 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:14:07.173 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'The images match the prompt well, showing a large rabbit with red hair holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 11:14:07.174 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'The images match the prompt well, showing a large rabbit with red hair holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 11:14:07.174 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: rabbit, large, with red hair, holding a bazooka, full body, negative: blurry, low detail, bad anatomy, extra limbs, deformed face, deformed tail, repetitive foliage, tiling textures, lighting, shadows, background, eval: The images match the prompt well, showing a large rabbit with red hair holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.
2025-11-13 11:14:07.174 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:14:07.174 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:56:08.335 | INFO | __main__:call_choose_model:294 - [Gradio] Choosing model for prompt: cute rabbit holding a bazooka
2025-11-13 11:56:08.661 | INFO | __main__:call_choose_model:299 - [Gradio] LLM chose model: sd
2025-11-13 11:56:08.891 | INFO | __main__:_reset_ui_and_set_steps:308 - [Gradio] Resetting UI, setting steps to 28
2025-11-13 11:56:09.195 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: cute rabbit holding a bazooka
2025-11-13 11:56:09.196 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: cute rabbit holding a bazooka (using qwen2.5:7b)
2025-11-13 11:56:10.042 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'rabbit, cute, holding a bazooka, full body', 'negative': 'blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background'}
2025-11-13 11:56:10.259 | INFO | __main__:route_generation:328 - [Gradio] Routing generation to model: sd with 28 steps
2025-11-13 11:56:10.260 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 28 inference steps...
2025-11-13 11:56:10.260 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 28 steps...
2025-11-13 11:56:22.528 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 11:56:22.555 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_wjsv9iid for Trellis step
2025-11-13 11:56:22.556 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 11:56:22.973 | INFO | __main__:route_post_processing:337 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-13 11:56:22.973 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: rabbit, cute, holding a bazooka, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval:
2025-11-13 11:56:22.974 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:56:22.974 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-13 11:56:22.974 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:56:24.338 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: rabbit, cute, holding a bazooka, full body, negative: extra limbs, deformed face, bad anatomy, shadows, lighting, background, unnatural pose, repetitive fur patterns, eval:
2025-11-13 11:56:24.338 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:56:24.338 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:56:28.873 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'the rabbit has an extra tail and a deformed body. The images are not consistent and some views have an extra tail. the face has low detail. the colors are unrealistic and unmatched.', 'negative_prompt': 'nsfw, explicit content, blurry, low detail, artifacts, geometry holes, extra tail, deformed body'}
2025-11-13 11:56:28.873 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'the rabbit has an extra tail and a deformed body. The images are not consistent and some views have an extra tail. the face has low detail. the colors are unrealistic and unmatched.', 'negative_prompt': 'nsfw, explicit content, blurry, low detail, artifacts, geometry holes, extra tail, deformed body'}
2025-11-13 11:56:28.874 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: rabbit, cute, holding a bazooka, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval: the rabbit has an extra tail and a deformed body. The images are not consistent and some views have an extra tail. the face has low detail. the colors are unrealistic and unmatched.
2025-11-13 11:56:28.874 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:56:28.874 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:57:18.392 | INFO | __main__:route_generation:328 - [Gradio] Routing generation to model: sd with 33 steps
2025-11-13 11:57:18.393 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 33 inference steps...
2025-11-13 11:57:18.393 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 33 steps...
2025-11-13 11:57:32.231 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 11:57:32.256 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_nkrkzio9 for Trellis step
2025-11-13 11:57:32.257 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 11:57:35.175 | INFO | __main__:route_post_processing:337 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-13 11:57:35.176 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: rabbit, cute, holding a bazooka, full body, realistic anatomy, negative: extra limbs, deformed face, deformed tail, low detail, artifacts, bad anatomy, unrealistic colors, inconsistent views, shadows, lighting, background, eval:
2025-11-13 11:57:35.178 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:57:35.180 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:57:35.181 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-13 11:57:36.543 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: rabbit, cute, holding a bazooka, full body, realistic anatomy, negative: extra limbs, deformed face, deformed tail, low detail, artifacts, bad anatomy, unrealistic colors, inconsistent views, grounded rabbit, unnatural shadows, blocky lighting, cluttered background, eval:
2025-11-13 11:57:36.543 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:57:36.543 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:57:40.995 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'the rabbit has an extra tail and a deformed body. The images are not consistent and some views have an extra tail. the face has low detail. the colors are unrealistic and unmatched.', 'negative_prompt': 'nsfw, explicit content, blurry, low detail, artifacts, geometry holes, extra tail, deformed body'}
2025-11-13 11:57:40.996 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'the rabbit has an extra tail and a deformed body. The images are not consistent and some views have an extra tail. the face has low detail. the colors are unrealistic and unmatched.', 'negative_prompt': 'nsfw, explicit content, blurry, low detail, artifacts, geometry holes, extra tail, deformed body'}
2025-11-13 11:57:40.996 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: rabbit, cute, holding a bazooka, full body, realistic anatomy, negative: extra limbs, deformed face, deformed tail, low detail, artifacts, bad anatomy, unrealistic colors, inconsistent views, shadows, lighting, background, eval: the rabbit has an extra tail and a deformed body. The images are not consistent and some views have an extra tail. the face has low detail. the colors are unrealistic and unmatched.
2025-11-13 11:57:40.996 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 11:57:40.997 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 11:58:36.926 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/views_nkrkzio9/trellis_output.zip
2025-11-13 11:58:36.968 | INFO | __main__:generate_3d:246 - [TRELLIS/API] Extracted ZIP to /tmp/views_nkrkzio9
2025-11-13 11:58:36.969 | INFO | __main__:generate_3d:249 - [TRELLIS/API-DEBUG] Files in /tmp/views_nkrkzio9: ['view_00.jpg', 'view_01.jpg', 'view_02.jpg', 'view_03.jpg', 'view_04.jpg', 'view_05.jpg', 'trellis_output.zip', 'mesh_ecdb251f-5370-4487-91f1-b3f258ace39b.glb', 'mesh_ecdb251f-5370-4487-91f1-b3f258ace39b.ply', 'preview_ecdb251f-5370-4487-91f1-b3f258ace39b.mp4', 'frame_ecdb251f-5370-4487-91f1-b3f258ace39b.jpg']
2025-11-13 11:58:36.969 | INFO | __main__:generate_3d:256 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_ecdb251f-5370-4487-91f1-b3f258ace39b.glb']
2025-11-13 11:58:36.969 | INFO | __main__:generate_3d:257 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_ecdb251f-5370-4487-91f1-b3f258ace39b.ply']
2025-11-13 11:58:36.969 | INFO | __main__:generate_3d:258 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_ecdb251f-5370-4487-91f1-b3f258ace39b.mp4']
2025-11-13 11:58:36.969 | INFO | __main__:generate_3d:259 - [TRELLIS/API-DEBUG] Found frame files: ['frame_ecdb251f-5370-4487-91f1-b3f258ace39b.jpg']
2025-11-13 11:58:36.969 | INFO | __main__:generate_3d:273 - [TRELLIS/API] Frame path found: /tmp/views_nkrkzio9/frame_ecdb251f-5370-4487-91f1-b3f258ace39b.jpg
2025-11-13 11:58:36.969 | INFO | __main__:generate_3d:277 - [TRELLIS/API] Assets found: GLB=/tmp/views_nkrkzio9/mesh_ecdb251f-5370-4487-91f1-b3f258ace39b.glb, PLY=/tmp/views_nkrkzio9/mesh_ecdb251f-5370-4487-91f1-b3f258ace39b.ply, MP4=/tmp/views_nkrkzio9/preview_ecdb251f-5370-4487-91f1-b3f258ace39b.mp4
2025-11-13 12:02:31.520 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/views_nkrkzio9/trellis_output.zip
2025-11-13 12:02:31.574 | INFO | __main__:generate_3d:246 - [TRELLIS/API] Extracted ZIP to /tmp/views_nkrkzio9
2025-11-13 12:02:31.575 | INFO | __main__:generate_3d:249 - [TRELLIS/API-DEBUG] Files in /tmp/views_nkrkzio9: ['view_00.jpg', 'view_01.jpg', 'view_02.jpg', 'view_03.jpg', 'view_04.jpg', 'view_05.jpg', 'trellis_output.zip', 'mesh_ecdb251f-5370-4487-91f1-b3f258ace39b.glb', 'mesh_ecdb251f-5370-4487-91f1-b3f258ace39b.ply', 'preview_ecdb251f-5370-4487-91f1-b3f258ace39b.mp4', 'frame_ecdb251f-5370-4487-91f1-b3f258ace39b.jpg', 'mesh_84db080e-c985-4650-ba42-34d5b1ddbbd5.glb', 'mesh_84db080e-c985-4650-ba42-34d5b1ddbbd5.ply', 'preview_84db080e-c985-4650-ba42-34d5b1ddbbd5.mp4', 'frame_84db080e-c985-4650-ba42-34d5b1ddbbd5.jpg']
2025-11-13 12:02:31.575 | INFO | __main__:generate_3d:256 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_ecdb251f-5370-4487-91f1-b3f258ace39b.glb', 'mesh_84db080e-c985-4650-ba42-34d5b1ddbbd5.glb']
2025-11-13 12:02:31.575 | INFO | __main__:generate_3d:257 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_ecdb251f-5370-4487-91f1-b3f258ace39b.ply', 'mesh_84db080e-c985-4650-ba42-34d5b1ddbbd5.ply']
2025-11-13 12:02:31.575 | INFO | __main__:generate_3d:258 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_ecdb251f-5370-4487-91f1-b3f258ace39b.mp4', 'preview_84db080e-c985-4650-ba42-34d5b1ddbbd5.mp4']
2025-11-13 12:02:31.575 | INFO | __main__:generate_3d:259 - [TRELLIS/API-DEBUG] Found frame files: ['frame_ecdb251f-5370-4487-91f1-b3f258ace39b.jpg', 'frame_84db080e-c985-4650-ba42-34d5b1ddbbd5.jpg']
2025-11-13 12:02:31.575 | INFO | __main__:generate_3d:273 - [TRELLIS/API] Frame path found: /tmp/views_nkrkzio9/frame_ecdb251f-5370-4487-91f1-b3f258ace39b.jpg
2025-11-13 12:02:31.575 | INFO | __main__:generate_3d:277 - [TRELLIS/API] Assets found: GLB=/tmp/views_nkrkzio9/mesh_ecdb251f-5370-4487-91f1-b3f258ace39b.glb, PLY=/tmp/views_nkrkzio9/mesh_ecdb251f-5370-4487-91f1-b3f258ace39b.ply, MP4=/tmp/views_nkrkzio9/preview_ecdb251f-5370-4487-91f1-b3f258ace39b.mp4
2025-11-13 12:12:14.068 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: monster bunny holding a bazooka
2025-11-13 12:12:16.535 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: sd
2025-11-13 12:12:16.725 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 28
2025-11-13 12:12:17.029 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: monster bunny holding a bazooka
2025-11-13 12:12:17.030 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: monster bunny holding a bazooka (using qwen2.5:7b)
2025-11-13 12:12:17.853 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'monster bunny, holding a bazooka, full body', 'negative': 'blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background'}
2025-11-13 12:12:18.043 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: sd with 28 steps
2025-11-13 12:12:18.044 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 28 inference steps...
2025-11-13 12:12:18.044 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 28 steps...
2025-11-13 12:12:29.993 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 12:12:30.029 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_yqo8ekkx for Trellis step
2025-11-13 12:12:30.029 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 12:12:30.454 | INFO | __main__:route_post_processing:352 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-13 12:12:30.454 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: monster bunny, holding a bazooka, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval:
2025-11-13 12:12:30.455 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 12:12:30.455 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-13 12:12:30.455 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 12:12:31.825 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: monster bunny, holding a bazooka, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, unnatural shadows, unrealistic lighting, cluttered background, eval:
2025-11-13 12:12:31.825 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 12:12:31.825 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 12:12:36.145 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'The images match the prompt well, showing a monster bunny holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 12:12:36.145 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'The images match the prompt well, showing a monster bunny holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 12:12:36.145 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: monster bunny, holding a bazooka, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval: The images match the prompt well, showing a monster bunny holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.
2025-11-13 12:12:36.145 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 12:12:36.145 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 12:15:00.569 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: sd with 33 steps
2025-11-13 12:15:00.569 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 33 inference steps...
2025-11-13 12:15:00.569 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 33 steps...
2025-11-13 12:15:14.414 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 12:15:14.463 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_r07ynl_h for Trellis step
2025-11-13 12:15:14.463 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 12:15:14.978 | INFO | __main__:route_post_processing:352 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-13 12:15:14.978 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: monster bunny, holding a bazooka, full body, negative: blurry, low detail, bad anatomy, extra limbs, deformed face, deformed tail, repetitive textures, shadows, lighting, background, eval:
2025-11-13 12:15:14.980 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-13 12:15:14.980 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 12:15:14.983 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 12:15:16.094 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: monster bunny, holding a bazooka, full body, negative: deformed anatomy, repetitive textures, unnatural colors, blocky water surface, flat water, over-saturated colors, shadows, lighting, background, eval:
2025-11-13 12:15:16.095 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 12:15:16.097 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 12:15:20.521 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'The images match the prompt well, showing a monster bunny holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 12:15:20.522 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'The images match the prompt well, showing a monster bunny holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 12:15:20.522 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: monster bunny, holding a bazooka, full body, negative: blurry, low detail, bad anatomy, extra limbs, deformed face, deformed tail, repetitive textures, shadows, lighting, background, eval: The images match the prompt well, showing a monster bunny holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.
2025-11-13 12:15:20.522 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 12:15:20.522 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 12:15:24.462 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: sd with 33 steps
2025-11-13 12:15:24.463 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 33 inference steps...
2025-11-13 12:15:24.463 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 33 steps...
2025-11-13 12:15:39.499 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 12:15:39.528 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_5t9jmffn for Trellis step
2025-11-13 12:15:39.529 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 12:15:39.937 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: monster bunny, holding a bazooka, full body, negative: blurry, low detail, bad anatomy, extra limbs, deformed face, deformed tail, repetitive textures, eval:
2025-11-13 12:15:39.937 | INFO | __main__:route_post_processing:352 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-13 12:15:39.938 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 12:15:39.941 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 12:15:39.942 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-13 12:15:41.105 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: monster bunny, holding a bazooka, full body, negative: blurry, low detail, bad anatomy, extra limbs, deformed face, deformed tail, repetitive textures, unnatural colors, blocky water surface, lighting inconsistencies, eval:
2025-11-13 12:15:41.105 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 12:15:41.105 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 12:15:45.373 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'The images match the prompt well, showing a monster bunny holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 12:15:45.374 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'The images match the prompt well, showing a monster bunny holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 12:15:45.374 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: monster bunny, holding a bazooka, full body, negative: blurry, low detail, bad anatomy, extra limbs, deformed face, deformed tail, repetitive textures, eval: The images match the prompt well, showing a monster bunny holding a bazooka. The images are consistent across different angles, and there are no visible issues or artifacts.
2025-11-13 12:15:45.374 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 12:15:45.375 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 12:16:30.080 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/views_5t9jmffn/trellis_output.zip
2025-11-13 12:16:30.080 | INFO | __main__:generate_3d:243 - [Gradio/Fix] Cleaning old mesh files from /tmp/views_5t9jmffn
2025-11-13 12:16:30.134 | INFO | __main__:generate_3d:261 - [TRELLIS/API] Extracted ZIP to /tmp/views_5t9jmffn
2025-11-13 12:16:30.134 | INFO | __main__:generate_3d:264 - [TRELLIS/API-DEBUG] Files in /tmp/views_5t9jmffn: ['view_00.jpg', 'view_01.jpg', 'view_02.jpg', 'view_03.jpg', 'view_04.jpg', 'view_05.jpg', 'trellis_output.zip', 'mesh_861cab4b-af5d-4962-9158-bba265380968.glb', 'mesh_861cab4b-af5d-4962-9158-bba265380968.ply', 'preview_861cab4b-af5d-4962-9158-bba265380968.mp4', 'frame_861cab4b-af5d-4962-9158-bba265380968.jpg']
2025-11-13 12:16:30.134 | INFO | __main__:generate_3d:271 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_861cab4b-af5d-4962-9158-bba265380968.glb']
2025-11-13 12:16:30.134 | INFO | __main__:generate_3d:272 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_861cab4b-af5d-4962-9158-bba265380968.ply']
2025-11-13 12:16:30.134 | INFO | __main__:generate_3d:273 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_861cab4b-af5d-4962-9158-bba265380968.mp4']
2025-11-13 12:16:30.134 | INFO | __main__:generate_3d:274 - [TRELLIS/API-DEBUG] Found frame files: ['frame_861cab4b-af5d-4962-9158-bba265380968.jpg']
2025-11-13 12:16:30.134 | INFO | __main__:generate_3d:288 - [TRELLIS/API] Frame path found: /tmp/views_5t9jmffn/frame_861cab4b-af5d-4962-9158-bba265380968.jpg
2025-11-13 12:16:30.134 | INFO | __main__:generate_3d:292 - [TRELLIS/API] Assets found: GLB=/tmp/views_5t9jmffn/mesh_861cab4b-af5d-4962-9158-bba265380968.glb, PLY=/tmp/views_5t9jmffn/mesh_861cab4b-af5d-4962-9158-bba265380968.ply, MP4=/tmp/views_5t9jmffn/preview_861cab4b-af5d-4962-9158-bba265380968.mp4
2025-11-13 14:23:51.898 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: two floors house catching on fire
2025-11-13 14:23:52.231 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: flux
2025-11-13 14:23:52.454 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 10
2025-11-13 14:23:52.807 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: two floors house catching on fire
2025-11-13 14:23:52.807 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: two floors house catching on fire (using qwen2.5:7b)
2025-11-13 14:23:53.595 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'house, two floors, catching on fire, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-13 14:23:53.777 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 14:23:53.777 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 14:23:53.777 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 14:23:59.000 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 14:23:59.009 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_hyk9p5rj for feedback
2025-11-13 14:23:59.009 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 14:23:59.434 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, two floors, catching on fire, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-13 14:23:59.435 | INFO | __main__:route_post_processing:368 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 14:23:59.435 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 14:23:59.435 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 14:24:00.804 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, two floors, catching on fire, full view, negative: bad proportions, deformed parts, floating objects, harsh shadows, unrealistic lighting, blocky textures, background distractions, duplicate elements, eval:
2025-11-13 14:24:00.805 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 14:24:00.805 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 14:24:27.938 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: flooded 3 floors house
2025-11-13 14:24:28.270 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: flux
2025-11-13 14:24:28.497 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 10
2025-11-13 14:24:28.800 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: flooded 3 floors house
2025-11-13 14:24:28.800 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: flooded 3 floors house (using qwen2.5:7b)
2025-11-13 14:24:29.600 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'house, flooded, 3 floors, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-13 14:24:29.814 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 14:24:29.814 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 14:24:29.814 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 14:24:34.724 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 14:24:34.733 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_00xw7sp_ for feedback
2025-11-13 14:24:34.733 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 14:24:35.157 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, flooded, 3 floors, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-13 14:24:35.157 | INFO | __main__:route_post_processing:368 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 14:24:35.158 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 14:24:35.159 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 14:24:36.374 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, flooded, 3 floors, full view, negative: low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background, eval:
2025-11-13 14:24:36.375 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 14:24:36.375 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 14:25:48.399 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/flux_view_00xw7sp_/trellis_output.zip
2025-11-13 14:25:48.400 | INFO | __main__:generate_3d:243 - [Gradio/Fix] Cleaning old mesh files from /tmp/flux_view_00xw7sp_
2025-11-13 14:25:48.493 | INFO | __main__:generate_3d:261 - [TRELLIS/API] Extracted ZIP to /tmp/flux_view_00xw7sp_
2025-11-13 14:25:48.493 | INFO | __main__:generate_3d:264 - [TRELLIS/API-DEBUG] Files in /tmp/flux_view_00xw7sp_: ['view_00.jpg', 'trellis_output.zip', 'mesh_76fc8da3-b7cb-4c25-afa1-724e96b49357.glb', 'mesh_76fc8da3-b7cb-4c25-afa1-724e96b49357.ply', 'preview_76fc8da3-b7cb-4c25-afa1-724e96b49357.mp4', 'frame_76fc8da3-b7cb-4c25-afa1-724e96b49357.jpg']
2025-11-13 14:25:48.493 | INFO | __main__:generate_3d:271 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_76fc8da3-b7cb-4c25-afa1-724e96b49357.glb']
2025-11-13 14:25:48.493 | INFO | __main__:generate_3d:272 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_76fc8da3-b7cb-4c25-afa1-724e96b49357.ply']
2025-11-13 14:25:48.493 | INFO | __main__:generate_3d:273 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_76fc8da3-b7cb-4c25-afa1-724e96b49357.mp4']
2025-11-13 14:25:48.493 | INFO | __main__:generate_3d:274 - [TRELLIS/API-DEBUG] Found frame files: ['frame_76fc8da3-b7cb-4c25-afa1-724e96b49357.jpg']
2025-11-13 14:25:48.493 | INFO | __main__:generate_3d:288 - [TRELLIS/API] Frame path found: /tmp/flux_view_00xw7sp_/frame_76fc8da3-b7cb-4c25-afa1-724e96b49357.jpg
2025-11-13 14:25:48.493 | INFO | __main__:generate_3d:292 - [TRELLIS/API] Assets found: GLB=/tmp/flux_view_00xw7sp_/mesh_76fc8da3-b7cb-4c25-afa1-724e96b49357.glb, PLY=/tmp/flux_view_00xw7sp_/mesh_76fc8da3-b7cb-4c25-afa1-724e96b49357.ply, MP4=/tmp/flux_view_00xw7sp_/preview_76fc8da3-b7cb-4c25-afa1-724e96b49357.mp4
2025-11-13 15:26:59.915 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 15:26:59.916 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 15:26:59.916 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 15:27:04.936 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 15:27:04.955 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_lyft4rov for feedback
2025-11-13 15:27:04.956 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 15:27:05.379 | INFO | __main__:route_post_processing:368 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 15:27:05.379 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, flooded, 3 floors, negative: low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background, eval:
2025-11-13 15:27:05.380 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:27:05.381 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:27:08.982 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, flooded, 3 floors, negative: artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background, blocky water surface, eval:
2025-11-13 15:27:08.982 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:27:08.983 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:27:17.148 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: house cracked in half
2025-11-13 15:27:17.485 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: flux
2025-11-13 15:27:17.706 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 10
2025-11-13 15:27:18.059 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: house cracked in half
2025-11-13 15:27:18.059 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: house cracked in half (using qwen2.5:7b)
2025-11-13 15:27:18.843 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'house, cracked in half, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-13 15:27:19.082 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 15:27:19.083 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 15:27:19.083 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 15:27:23.996 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 15:27:24.006 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_b_m42gje for feedback
2025-11-13 15:27:24.006 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 15:27:24.430 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, cracked in half, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-13 15:27:24.430 | INFO | __main__:route_post_processing:368 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 15:27:24.588 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:27:24.590 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:27:25.800 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, cracked in half, full view, negative: bad proportions, deformed parts, artifacts, low detail, blurry, floating objects, unnatural shadows, poor lighting, cluttered background, eval:
2025-11-13 15:27:25.802 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:27:25.802 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:28:47.052 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 15:28:47.052 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 15:28:47.053 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 15:28:51.974 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 15:28:51.981 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_q17y4ca0 for feedback
2025-11-13 15:28:51.982 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 15:28:58.168 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, cracked in half, full view, negative: bad proportions, deformed parts, artifacts, low detail, blurry, floating objects, unnatural shadows, poor lighting, eval:
2025-11-13 15:28:58.168 | INFO | __main__:route_post_processing:368 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 15:28:58.171 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:28:58.175 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:28:59.130 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, cracked in half, full view, negative: bad proportions, deformed parts, artifacts, low detail, blurry, floating objects, unnatural shadows, poor lighting, overcast sky, eval:
2025-11-13 15:28:59.130 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:28:59.130 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:29:03.317 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 15:29:03.317 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 15:29:03.317 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 15:29:08.258 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 15:29:08.272 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_dycv5eh3 for feedback
2025-11-13 15:29:08.272 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 15:29:12.568 | INFO | __main__:route_post_processing:368 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 15:29:12.568 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, cracked in half, full view, negative: bad proportions, deformed parts, artifacts, low detail, blurry, floating objects, unnatural shadows, poor lighting, eval:
2025-11-13 15:29:12.572 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:29:12.573 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:29:13.585 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: house, cracked in half, full view, negative: bad proportions, deformed parts, artifacts, low detail, blurry, floating objects, unnatural shadows, poor lighting, inconsistent cracks, eval:
2025-11-13 15:29:13.585 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:29:13.585 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:29:18.690 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: gas station
2025-11-13 15:29:19.018 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: flux
2025-11-13 15:29:19.247 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 10
2025-11-13 15:29:19.500 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: gas station
2025-11-13 15:29:19.501 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: gas station (using qwen2.5:7b)
2025-11-13 15:29:20.278 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'gas station, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-13 15:29:20.468 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 15:29:20.468 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 15:29:20.469 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 15:29:25.387 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 15:29:25.405 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_c7237m73 for feedback
2025-11-13 15:29:25.406 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 15:29:25.772 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: gas station, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-13 15:29:25.773 | INFO | __main__:route_post_processing:368 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 15:29:25.773 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:29:25.775 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:29:26.990 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: gas station, full view, negative: bad proportions, deformed parts, artifacts, low detail, floating objects, unnatural shadows, poor lighting, cluttered background, eval:
2025-11-13 15:29:26.991 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:29:26.991 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:29:57.300 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: gas station catching fire
2025-11-13 15:29:57.629 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: flux
2025-11-13 15:29:57.859 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 10
2025-11-13 15:29:58.212 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: gas station catching fire
2025-11-13 15:29:58.212 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: gas station catching fire (using qwen2.5:7b)
2025-11-13 15:29:58.972 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'gas station, catching fire, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-13 15:29:59.173 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 15:29:59.173 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 15:29:59.174 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 15:30:04.088 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 15:30:04.096 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_cr37k6_2 for feedback
2025-11-13 15:30:04.096 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 15:30:04.515 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: gas station, catching fire, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-13 15:30:04.516 | INFO | __main__:route_post_processing:368 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 15:30:04.516 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:30:04.519 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:30:05.936 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: gas station, catching fire, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, incorrect lighting, cluttered background, eval:
2025-11-13 15:30:05.937 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:30:05.937 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:30:20.294 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: dog knight
2025-11-13 15:30:20.607 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: sd
2025-11-13 15:30:20.800 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 28
2025-11-13 15:30:21.103 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: dog knight
2025-11-13 15:30:21.103 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: dog knight (using qwen2.5:7b)
2025-11-13 15:30:21.888 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'dog, knight, full body', 'negative': 'blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background'}
2025-11-13 15:30:22.070 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: sd with 28 steps
2025-11-13 15:30:22.070 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 28 inference steps...
2025-11-13 15:30:22.070 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 28 steps...
2025-11-13 15:30:33.992 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 15:30:34.019 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_1x9smcwl for Trellis step
2025-11-13 15:30:34.020 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 15:30:34.446 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: dog, knight, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval:
2025-11-13 15:30:34.446 | INFO | __main__:route_post_processing:352 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-13 15:30:34.449 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:30:34.451 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:30:34.451 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-13 15:30:38.916 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'The images match the prompt well, showing a dog knight from different angles. The images are consistent across the views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 15:30:38.917 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'The images match the prompt well, showing a dog knight from different angles. The images are consistent across the views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 15:30:38.917 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: dog, knight, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval: The images match the prompt well, showing a dog knight from different angles. The images are consistent across the views, and there are no visible issues or artifacts.
2025-11-13 15:30:38.919 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:30:38.919 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:30:45.620 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: dog, full body knight, negative: artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval:
2025-11-13 15:30:45.625 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:30:45.626 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:40:17.507 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: cat knight
2025-11-13 15:40:20.036 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: sd
2025-11-13 15:40:34.330 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 28
2025-11-13 15:40:34.637 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: cat knight
2025-11-13 15:40:34.637 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: cat knight (using qwen2.5:7b)
2025-11-13 15:40:35.497 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'cat, knight, full body', 'negative': 'blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background'}
2025-11-13 15:40:35.703 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: sd with 28 steps
2025-11-13 15:40:35.704 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 28 inference steps...
2025-11-13 15:40:35.704 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 28 steps...
2025-11-13 15:40:47.558 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 15:40:47.617 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_3c8430gf for Trellis step
2025-11-13 15:40:47.618 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 15:40:48.444 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: cat, knight, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval:
2025-11-13 15:40:48.444 | INFO | __main__:route_post_processing:352 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-13 15:40:48.449 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:40:48.452 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:40:48.453 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-13 15:40:49.710 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: cat, knight, full body, negative: artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval:
2025-11-13 15:40:49.713 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:40:49.718 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:40:54.140 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': "The images match the prompt well, showing a cat in a knight's outfit from different angles. The images are consistent across the views, and there are no visible issues or artifacts.", 'negative_prompt': 'none'}
2025-11-13 15:40:54.140 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': "The images match the prompt well, showing a cat in a knight's outfit from different angles. The images are consistent across the views, and there are no visible issues or artifacts.", 'negative_prompt': 'none'}
2025-11-13 15:40:54.140 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: cat, knight, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval: The images match the prompt well, showing a cat in a knight's outfit from different angles. The images are consistent across the views, and there are no visible issues or artifacts.
2025-11-13 15:40:54.140 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:40:54.140 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:54:15.685 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: river between trees
2025-11-13 15:54:18.179 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: flux
2025-11-13 15:54:18.370 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 10
2025-11-13 15:54:18.673 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: river between trees
2025-11-13 15:54:18.673 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: river between trees (using qwen2.5:7b)
2025-11-13 15:54:19.452 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'river, between trees, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-13 15:54:19.684 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 15:54:19.684 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 15:54:19.684 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 15:54:24.610 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 15:54:24.636 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_0oyw1ize for feedback
2025-11-13 15:54:24.636 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 15:54:25.082 | INFO | __main__:route_post_processing:368 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 15:54:25.082 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: river, between trees, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-13 15:54:25.083 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:54:25.084 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:54:26.452 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: river, between trees, full view, negative: bad proportions, artifacts, deformed parts, floating objects, unnatural shadows, poor lighting, blocky water surface, background, eval:
2025-11-13 15:54:26.452 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 15:54:26.453 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 15:55:23.766 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/flux_view_0oyw1ize/trellis_output.zip
2025-11-13 15:55:23.767 | INFO | __main__:generate_3d:243 - [Gradio/Fix] Cleaning old mesh files from /tmp/flux_view_0oyw1ize
2025-11-13 15:55:23.834 | INFO | __main__:generate_3d:261 - [TRELLIS/API] Extracted ZIP to /tmp/flux_view_0oyw1ize
2025-11-13 15:55:23.834 | INFO | __main__:generate_3d:264 - [TRELLIS/API-DEBUG] Files in /tmp/flux_view_0oyw1ize: ['view_00.jpg', 'trellis_output.zip', 'mesh_253e756c-bfe7-445f-b1fd-558e60cf8cb8.glb', 'mesh_253e756c-bfe7-445f-b1fd-558e60cf8cb8.ply', 'preview_253e756c-bfe7-445f-b1fd-558e60cf8cb8.mp4', 'frame_253e756c-bfe7-445f-b1fd-558e60cf8cb8.jpg']
2025-11-13 15:55:23.834 | INFO | __main__:generate_3d:271 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_253e756c-bfe7-445f-b1fd-558e60cf8cb8.glb']
2025-11-13 15:55:23.834 | INFO | __main__:generate_3d:272 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_253e756c-bfe7-445f-b1fd-558e60cf8cb8.ply']
2025-11-13 15:55:23.835 | INFO | __main__:generate_3d:273 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_253e756c-bfe7-445f-b1fd-558e60cf8cb8.mp4']
2025-11-13 15:55:23.835 | INFO | __main__:generate_3d:274 - [TRELLIS/API-DEBUG] Found frame files: ['frame_253e756c-bfe7-445f-b1fd-558e60cf8cb8.jpg']
2025-11-13 15:55:23.835 | INFO | __main__:generate_3d:288 - [TRELLIS/API] Frame path found: /tmp/flux_view_0oyw1ize/frame_253e756c-bfe7-445f-b1fd-558e60cf8cb8.jpg
2025-11-13 15:55:23.835 | INFO | __main__:generate_3d:292 - [TRELLIS/API] Assets found: GLB=/tmp/flux_view_0oyw1ize/mesh_253e756c-bfe7-445f-b1fd-558e60cf8cb8.glb, PLY=/tmp/flux_view_0oyw1ize/mesh_253e756c-bfe7-445f-b1fd-558e60cf8cb8.ply, MP4=/tmp/flux_view_0oyw1ize/preview_253e756c-bfe7-445f-b1fd-558e60cf8cb8.mp4
2025-11-13 17:52:55.409 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: cute dog
2025-11-13 17:52:57.883 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: sd
2025-11-13 17:53:00.814 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 28
2025-11-13 17:53:01.117 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: cute dog
2025-11-13 17:53:01.118 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: cute dog (using qwen2.5:7b)
2025-11-13 17:53:01.946 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'dog, cute, full body', 'negative': 'blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background'}
2025-11-13 17:53:02.130 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: sd with 28 steps
2025-11-13 17:53:02.130 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 28 inference steps...
2025-11-13 17:53:02.131 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 28 steps...
2025-11-13 17:53:14.034 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 17:53:14.060 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_k_p3zqvd for Trellis step
2025-11-13 17:53:14.060 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 17:53:14.481 | INFO | __main__:route_post_processing:352 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-13 17:53:14.482 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: dog, cute, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval:
2025-11-13 17:53:14.483 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-13 17:53:14.483 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 17:53:14.489 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 17:53:15.800 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: dog, cute, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, unnatural shadows, unrealistic lighting, cluttered background, eval:
2025-11-13 17:53:15.804 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 17:53:15.805 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 17:53:19.990 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'the images match the prompt well. The images are consistent across different angles, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 17:53:19.991 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'the images match the prompt well. The images are consistent across different angles, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 17:53:19.991 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: dog, cute, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval: the images match the prompt well. The images are consistent across different angles, and there are no visible issues or artifacts.
2025-11-13 17:53:19.991 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 17:53:19.991 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 17:54:50.395 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/views_k_p3zqvd/trellis_output.zip
2025-11-13 17:54:50.396 | INFO | __main__:generate_3d:243 - [Gradio/Fix] Cleaning old mesh files from /tmp/views_k_p3zqvd
2025-11-13 17:54:50.486 | INFO | __main__:generate_3d:261 - [TRELLIS/API] Extracted ZIP to /tmp/views_k_p3zqvd
2025-11-13 17:54:50.486 | INFO | __main__:generate_3d:264 - [TRELLIS/API-DEBUG] Files in /tmp/views_k_p3zqvd: ['view_00.jpg', 'view_01.jpg', 'view_02.jpg', 'view_03.jpg', 'view_04.jpg', 'view_05.jpg', 'trellis_output.zip', 'mesh_fa482743-968f-4a2d-b041-89c7e3df88bb.glb', 'mesh_fa482743-968f-4a2d-b041-89c7e3df88bb.ply', 'preview_fa482743-968f-4a2d-b041-89c7e3df88bb.mp4', 'frame_fa482743-968f-4a2d-b041-89c7e3df88bb.jpg']
2025-11-13 17:54:50.486 | INFO | __main__:generate_3d:271 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_fa482743-968f-4a2d-b041-89c7e3df88bb.glb']
2025-11-13 17:54:50.486 | INFO | __main__:generate_3d:272 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_fa482743-968f-4a2d-b041-89c7e3df88bb.ply']
2025-11-13 17:54:50.486 | INFO | __main__:generate_3d:273 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_fa482743-968f-4a2d-b041-89c7e3df88bb.mp4']
2025-11-13 17:54:50.486 | INFO | __main__:generate_3d:274 - [TRELLIS/API-DEBUG] Found frame files: ['frame_fa482743-968f-4a2d-b041-89c7e3df88bb.jpg']
2025-11-13 17:54:50.486 | INFO | __main__:generate_3d:288 - [TRELLIS/API] Frame path found: /tmp/views_k_p3zqvd/frame_fa482743-968f-4a2d-b041-89c7e3df88bb.jpg
2025-11-13 17:54:50.486 | INFO | __main__:generate_3d:292 - [TRELLIS/API] Assets found: GLB=/tmp/views_k_p3zqvd/mesh_fa482743-968f-4a2d-b041-89c7e3df88bb.glb, PLY=/tmp/views_k_p3zqvd/mesh_fa482743-968f-4a2d-b041-89c7e3df88bb.ply, MP4=/tmp/views_k_p3zqvd/preview_fa482743-968f-4a2d-b041-89c7e3df88bb.mp4
2025-11-13 17:57:00.040 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: gas station
2025-11-13 17:57:00.364 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: flux
2025-11-13 17:57:00.596 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 10
2025-11-13 17:57:00.900 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: gas station
2025-11-13 17:57:00.900 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: gas station (using qwen2.5:7b)
2025-11-13 17:57:01.678 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'gas station, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-13 17:57:01.862 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 17:57:01.862 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 17:57:01.862 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 17:57:06.813 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 17:57:06.820 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_g4txq_ek for feedback
2025-11-13 17:57:06.821 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 17:57:07.241 | INFO | __main__:route_post_processing:368 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 17:57:07.242 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: gas station, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-13 17:57:07.242 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 17:57:07.244 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 17:57:08.407 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: gas station, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background, eval:
2025-11-13 17:57:08.407 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 17:57:08.407 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 17:58:11.150 | INFO | ollama_chat:continue_chat_with_feedback:125 - [CHAT] User feedback: on fire
2025-11-13 17:58:11.151 | INFO | ollama_chat:continue_chat_with_feedback:126 - [CHAT] Messages: [{'role': 'system', 'content': 'You are a prompt engineer and image reviewer for text-to-3D/multiview generation.\n\nYou will receive:\n- an image avaliation (text),\n- the current positive prompt (refined),\n- the current negative prompt.\nAvaliation:\n- Compare the image quality to the prompts.\n- You can be critical but not too much in your evaluation. If the images have minor issues but overall look good, say so. If they have major problems, point them out clearly. Only give constructive feedback.\n- If the overall quality is good, you don\'t need to list every small issue and the negative prompt can be the same negative prompt received.\n- The bigger problems are: geometry holes, extra limbs (if the prompt doesn\'t call for it), deformed face, artifacts, lighting, shadows, background.\n- Example: if the input is "a cat", the images should look like a real cat from different angles. If the cat has extra limbs, deformed face, a extra tail, say so.\n\nBehaviors:\n- On START: return the three fields in the strict JSON format (you may lightly polish wording).\n- On FEEDBACK: YOU MUST APPLY the client\'s requests by EDITING the prompts.\n - If the user asks to ADD or EMPHASIZE something > add concise tokens to the POSITIVE.\n - If the user asks to REDUCE/REMOVE or MITIGATE errors/artifacts > add/adjust constraints in the NEGATIVE (and optionally remove tokens from POSITIVE if conflicting).\n - If the user complain about something he saw, interpret it and add/remove/modify tokens accordingly on negative/positive. Example: it doesn\'t look natural > add "unnatural colors" to negative; it has too many trees > add "excessive trees" to negative; the dragon is not flying how i asked on the prompt > add "flying dragon" to positive and "grounded dragon" to negative.\n - Never return unchanged prompts when edits were requested.\n- Never ignore the user\'s requests. Insert the changes based on the user\'s feedback and don\'t ignore the user\'s requests.\n- Specify the chacteristics in the prompts using concise, specific tokens (no vague or generic terms). For example, if it says "cat with full orange body", specify "orange body", "orange members", "orange head", "orange tail", etc. But you don\'t need to be exhaustive.\n\nPositive rules:\n- Object-first (singular), concise attributes, preserve multi-word spatial relations; comma-separated.\n- If the user wants to add something, add it to the positive prompt.\n- Use concrete, visual tokens (avoid vague phrasing).\n- Examples:\n * "more water" > add "dominant lake", "expanded water surface"\n * "increase detail on water surface" > add "detailed water surface"\n * "natural look" > add "natural colors", "moderate saturation"\n * "center the lake" > add "central lake", "lake in center"\n * "more green" > add "lush greenery", "vibrant foliage"\n\n* "less trees" > remove objects or caracteristics if present; do NOT add to positive. Mitigate or remove in negative.\n\nNegative rules:\n- 6-8 items MAX (do not exceed); focus on artifacts/geometry/texture; comma-separated.\n-If the user wants to mitigate/reduce/remove something, add it to the negative prompt.\n- Map mitigation requests to negatives, examle.:\n * less trees > "excessive trees", "overcrowded foliage", "duplicate trees"\n * avoid repetitive foliage > "repetitive foliage", "tiling textures"\n * flat water > "flat water", "blocky water surface", "unrealistic reflections"\n * reduce saturation > "over-saturated colors"\n * remove background > keep "background" in negative\n- ALWAYS append (even beyond the 8 cap): lighting, shadows, background.\n\n- If the user complaint about something he saw or missed on the image, interpret it and add/remove/modify tokens accordingly on negative/positive.\nAlways have certain if a caracteristic should be in positive or negative. For example:\n- "more trees" > add to positive\n- "less trees" > mitigate/remove in negative (do NOT add to positive)\n- "more detail" > add to positive\n- "reduce detail" > mitigate/remove in negative (do NOT add to positive)\n- "natural look" > add to positive\n- "avoid artifacts" > add to negative\n\nConversation rule:\n- Always include a short 1-2 sentence natural-language "message" addressing the user. On START: invite edits. On FEEDBACK: acknowledge changes and ask if more tweaks are needed. Keep it concise and friendly.\n\nSTRICT OUTPUT (JSON ONLY; no extra text):\n{\n "avaliation": "<1-4 sentences>",\n "refined": "<positive prompt, comma-separated>",\n "negative": "<negative prompt, comma-separated; 6-8 items MAX + lighting, shadows, background>",\n "message": "<1-2 short sentences to the user>"\n}'}, {'role': 'user', 'content': 'START. Use the provided avaliation and current prompts; return ONLY the strict JSON.\n\n- avaliation: \n- Current positive: gas station, full view\n- Current negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background\n'}, {'role': 'assistant', 'content': '{"avaliation": "The current prompts are good but could benefit from more details and specific attributes.", "refined": "gas station, full view, modern design, well-lit, clear signage", "negative": "blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background", "message": "I\'ve added some specific attributes to the positive prompt. Let me know if you need any further adjustments!"}'}]
2025-11-13 17:58:11.151 | INFO | ollama_chat:continue_chat_with_feedback:148 - [CHAT] continue_chat_with_feedback (using qwen2.5:7b)
2025-11-13 17:58:11.151 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 17:58:15.189 | INFO | ollama_chat:continue_chat_with_feedback:125 - [CHAT] User feedback: ok
2025-11-13 17:58:15.189 | INFO | ollama_chat:continue_chat_with_feedback:126 - [CHAT] Messages: [{'role': 'system', 'content': 'You are a prompt engineer and image reviewer for text-to-3D/multiview generation.\n\nYou will receive:\n- an image avaliation (text),\n- the current positive prompt (refined),\n- the current negative prompt.\nAvaliation:\n- Compare the image quality to the prompts.\n- You can be critical but not too much in your evaluation. If the images have minor issues but overall look good, say so. If they have major problems, point them out clearly. Only give constructive feedback.\n- If the overall quality is good, you don\'t need to list every small issue and the negative prompt can be the same negative prompt received.\n- The bigger problems are: geometry holes, extra limbs (if the prompt doesn\'t call for it), deformed face, artifacts, lighting, shadows, background.\n- Example: if the input is "a cat", the images should look like a real cat from different angles. If the cat has extra limbs, deformed face, a extra tail, say so.\n\nBehaviors:\n- On START: return the three fields in the strict JSON format (you may lightly polish wording).\n- On FEEDBACK: YOU MUST APPLY the client\'s requests by EDITING the prompts.\n - If the user asks to ADD or EMPHASIZE something > add concise tokens to the POSITIVE.\n - If the user asks to REDUCE/REMOVE or MITIGATE errors/artifacts > add/adjust constraints in the NEGATIVE (and optionally remove tokens from POSITIVE if conflicting).\n - If the user complain about something he saw, interpret it and add/remove/modify tokens accordingly on negative/positive. Example: it doesn\'t look natural > add "unnatural colors" to negative; it has too many trees > add "excessive trees" to negative; the dragon is not flying how i asked on the prompt > add "flying dragon" to positive and "grounded dragon" to negative.\n - Never return unchanged prompts when edits were requested.\n- Never ignore the user\'s requests. Insert the changes based on the user\'s feedback and don\'t ignore the user\'s requests.\n- Specify the chacteristics in the prompts using concise, specific tokens (no vague or generic terms). For example, if it says "cat with full orange body", specify "orange body", "orange members", "orange head", "orange tail", etc. But you don\'t need to be exhaustive.\n\nPositive rules:\n- Object-first (singular), concise attributes, preserve multi-word spatial relations; comma-separated.\n- If the user wants to add something, add it to the positive prompt.\n- Use concrete, visual tokens (avoid vague phrasing).\n- Examples:\n * "more water" > add "dominant lake", "expanded water surface"\n * "increase detail on water surface" > add "detailed water surface"\n * "natural look" > add "natural colors", "moderate saturation"\n * "center the lake" > add "central lake", "lake in center"\n * "more green" > add "lush greenery", "vibrant foliage"\n\n* "less trees" > remove objects or caracteristics if present; do NOT add to positive. Mitigate or remove in negative.\n\nNegative rules:\n- 6-8 items MAX (do not exceed); focus on artifacts/geometry/texture; comma-separated.\n-If the user wants to mitigate/reduce/remove something, add it to the negative prompt.\n- Map mitigation requests to negatives, examle.:\n * less trees > "excessive trees", "overcrowded foliage", "duplicate trees"\n * avoid repetitive foliage > "repetitive foliage", "tiling textures"\n * flat water > "flat water", "blocky water surface", "unrealistic reflections"\n * reduce saturation > "over-saturated colors"\n * remove background > keep "background" in negative\n- ALWAYS append (even beyond the 8 cap): lighting, shadows, background.\n\n- If the user complaint about something he saw or missed on the image, interpret it and add/remove/modify tokens accordingly on negative/positive.\nAlways have certain if a caracteristic should be in positive or negative. For example:\n- "more trees" > add to positive\n- "less trees" > mitigate/remove in negative (do NOT add to positive)\n- "more detail" > add to positive\n- "reduce detail" > mitigate/remove in negative (do NOT add to positive)\n- "natural look" > add to positive\n- "avoid artifacts" > add to negative\n\nConversation rule:\n- Always include a short 1-2 sentence natural-language "message" addressing the user. On START: invite edits. On FEEDBACK: acknowledge changes and ask if more tweaks are needed. Keep it concise and friendly.\n\nSTRICT OUTPUT (JSON ONLY; no extra text):\n{\n "avaliation": "<1-4 sentences>",\n "refined": "<positive prompt, comma-separated>",\n "negative": "<negative prompt, comma-separated; 6-8 items MAX + lighting, shadows, background>",\n "message": "<1-2 short sentences to the user>"\n}'}, {'role': 'user', 'content': 'START. Use the provided avaliation and current prompts; return ONLY the strict JSON.\n\n- avaliation: \n- Current positive: gas station, full view\n- Current negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background\n'}, {'role': 'assistant', 'content': '{"avaliation": "The current prompts are good but could benefit from more details and specific attributes.", "refined": "gas station, full view, modern design, well-lit, clear signage", "negative": "blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background", "message": "I\'ve added some specific attributes to the positive prompt. Let me know if you need any further adjustments!"}'}, {'role': 'user', 'content': "FEEDBACK. Apply the client's requests by editing the prompts (add/remove/modify tokens). Return ONLY the strict JSON. Do not keep prompts unchanged if edits were requested.\n\n- Client feedback: on fire\n- Current avaliation: The current prompts are good but could benefit from more details and specific attributes.\n- Current positive: gas station, full view, modern design, well-lit, clear signage\n- Current negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background\n"}, {'role': 'assistant', 'content': '{"avaliation": "Added \'on fire\' to the positive prompt for more detail.", "refined": "gas station, full view, modern design, well-lit, clear signage, on fire", "negative": "blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background, excessive flames, incorrect fire effects", "message": "I\'ve added \'on fire\' to the positive prompt and mitigated potential issues with fire in the negative. Let me know if you need further adjustments!"}'}]
2025-11-13 17:58:18.719 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 10 inference steps...
2025-11-13 17:58:18.720 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 10 steps...
2025-11-13 17:58:23.425 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 17:58:23.454 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_u2ffouxx for Trellis step
2025-11-13 17:58:23.454 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 17:58:23.823 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: gas station, full view, modern design, well-lit, clear signage, on fire, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background, excessive flames, incorrect fire effects, eval:
2025-11-13 17:58:23.823 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 17:58:23.823 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 17:58:25.153 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: gas station, full view, modern design, well-lit, clear signage, burning, negative: artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background, excessive flames, incorrect fire effects, harsh shadows, over-saturated colors, background elements, eval:
2025-11-13 17:58:25.153 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 17:58:25.154 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 17:58:46.535 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: gas station
2025-11-13 17:58:46.845 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: flux
2025-11-13 17:58:47.093 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 10
2025-11-13 17:58:47.446 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: gas station
2025-11-13 17:58:47.447 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: gas station (using qwen2.5:7b)
2025-11-13 17:58:48.173 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'gas station, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-13 17:58:48.409 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 17:58:48.409 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 17:58:48.409 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 17:58:53.320 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 17:58:53.330 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_6yv_vgjz for feedback
2025-11-13 17:58:53.330 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 17:58:53.747 | INFO | __main__:route_post_processing:368 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 17:58:53.899 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: gas station, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-13 17:58:53.899 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 17:58:53.899 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 17:58:55.170 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: gas station, full view, negative: bad proportions, artifacts, deformed parts, floating objects, unnatural shadows, poor lighting, cluttered background, excessive textures, eval:
2025-11-13 17:58:55.176 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 17:58:55.177 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 19:21:51.860 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: green bike
2025-11-13 19:21:54.332 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: sd
2025-11-13 19:21:54.547 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 28
2025-11-13 19:21:54.899 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: green bike
2025-11-13 19:21:54.900 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: green bike (using qwen2.5:7b)
2025-11-13 19:21:55.620 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'bike, green, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-13 19:21:55.812 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: sd with 28 steps
2025-11-13 19:21:55.812 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 28 inference steps...
2025-11-13 19:21:55.812 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 28 steps...
2025-11-13 19:22:07.666 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 19:22:07.692 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_xic197c2 for Trellis step
2025-11-13 19:22:07.692 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 19:22:15.558 | INFO | __main__:route_post_processing:352 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-13 19:22:15.558 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: bike, green, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-13 19:22:15.559 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 19:22:15.559 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-13 19:22:15.559 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 19:22:16.886 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: bike, green, full view, negative: low detail, bad proportions, deformed parts, floating objects, unnatural shadows, poor lighting, background elements, repetitive textures, eval:
2025-11-13 19:22:16.886 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 19:22:16.886 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 19:22:21.243 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'The images match the prompt well, showing a green bike from different angles. The images are consistent across the views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 19:22:21.244 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'The images match the prompt well, showing a green bike from different angles. The images are consistent across the views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 19:22:21.244 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: bike, green, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval: The images match the prompt well, showing a green bike from different angles. The images are consistent across the views, and there are no visible issues or artifacts.
2025-11-13 19:22:21.244 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 19:22:21.245 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 19:23:09.141 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/views_xic197c2/trellis_output.zip
2025-11-13 19:23:09.142 | INFO | __main__:generate_3d:243 - [Gradio/Fix] Cleaning old mesh files from /tmp/views_xic197c2
2025-11-13 19:23:09.208 | INFO | __main__:generate_3d:261 - [TRELLIS/API] Extracted ZIP to /tmp/views_xic197c2
2025-11-13 19:23:09.208 | INFO | __main__:generate_3d:264 - [TRELLIS/API-DEBUG] Files in /tmp/views_xic197c2: ['view_00.jpg', 'view_01.jpg', 'view_02.jpg', 'view_03.jpg', 'view_04.jpg', 'view_05.jpg', 'trellis_output.zip', 'mesh_37027340-2ac8-456f-950a-78ada4a07dea.glb', 'mesh_37027340-2ac8-456f-950a-78ada4a07dea.ply', 'preview_37027340-2ac8-456f-950a-78ada4a07dea.mp4', 'frame_37027340-2ac8-456f-950a-78ada4a07dea.jpg']
2025-11-13 19:23:09.208 | INFO | __main__:generate_3d:271 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_37027340-2ac8-456f-950a-78ada4a07dea.glb']
2025-11-13 19:23:09.208 | INFO | __main__:generate_3d:272 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_37027340-2ac8-456f-950a-78ada4a07dea.ply']
2025-11-13 19:23:09.209 | INFO | __main__:generate_3d:273 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_37027340-2ac8-456f-950a-78ada4a07dea.mp4']
2025-11-13 19:23:09.209 | INFO | __main__:generate_3d:274 - [TRELLIS/API-DEBUG] Found frame files: ['frame_37027340-2ac8-456f-950a-78ada4a07dea.jpg']
2025-11-13 19:23:09.209 | INFO | __main__:generate_3d:288 - [TRELLIS/API] Frame path found: /tmp/views_xic197c2/frame_37027340-2ac8-456f-950a-78ada4a07dea.jpg
2025-11-13 19:23:09.209 | INFO | __main__:generate_3d:292 - [TRELLIS/API] Assets found: GLB=/tmp/views_xic197c2/mesh_37027340-2ac8-456f-950a-78ada4a07dea.glb, PLY=/tmp/views_xic197c2/mesh_37027340-2ac8-456f-950a-78ada4a07dea.ply, MP4=/tmp/views_xic197c2/preview_37027340-2ac8-456f-950a-78ada4a07dea.mp4
2025-11-13 19:44:41.309 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: river between trees
2025-11-13 19:44:43.846 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: flux
2025-11-13 19:44:44.046 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 10
2025-11-13 19:44:44.400 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: river between trees
2025-11-13 19:44:44.400 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: river between trees (using qwen2.5:7b)
2025-11-13 19:44:45.182 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'river, between trees, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-13 19:45:06.147 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 19:45:06.148 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 19:45:06.148 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 19:45:11.153 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 19:45:11.175 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_10ccuwew for feedback
2025-11-13 19:45:11.176 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 19:45:11.646 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: river, between trees, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-13 19:45:11.647 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 19:45:11.648 | INFO | __main__:route_post_processing:368 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 19:45:11.649 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 19:45:12.911 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: river, between trees, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, unnatural shadows, unrealistic lighting, cluttered background, eval:
2025-11-13 19:45:12.912 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 19:45:12.912 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 19:46:07.927 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/flux_view_10ccuwew/trellis_output.zip
2025-11-13 19:46:07.928 | INFO | __main__:generate_3d:243 - [Gradio/Fix] Cleaning old mesh files from /tmp/flux_view_10ccuwew
2025-11-13 19:46:08.016 | INFO | __main__:generate_3d:261 - [TRELLIS/API] Extracted ZIP to /tmp/flux_view_10ccuwew
2025-11-13 19:46:08.016 | INFO | __main__:generate_3d:264 - [TRELLIS/API-DEBUG] Files in /tmp/flux_view_10ccuwew: ['view_00.jpg', 'trellis_output.zip', 'mesh_5406073f-5c42-4e9a-b03d-f136dbd14931.glb', 'mesh_5406073f-5c42-4e9a-b03d-f136dbd14931.ply', 'preview_5406073f-5c42-4e9a-b03d-f136dbd14931.mp4', 'frame_5406073f-5c42-4e9a-b03d-f136dbd14931.jpg']
2025-11-13 19:46:08.017 | INFO | __main__:generate_3d:271 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_5406073f-5c42-4e9a-b03d-f136dbd14931.glb']
2025-11-13 19:46:08.017 | INFO | __main__:generate_3d:272 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_5406073f-5c42-4e9a-b03d-f136dbd14931.ply']
2025-11-13 19:46:08.017 | INFO | __main__:generate_3d:273 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_5406073f-5c42-4e9a-b03d-f136dbd14931.mp4']
2025-11-13 19:46:08.017 | INFO | __main__:generate_3d:274 - [TRELLIS/API-DEBUG] Found frame files: ['frame_5406073f-5c42-4e9a-b03d-f136dbd14931.jpg']
2025-11-13 19:46:08.017 | INFO | __main__:generate_3d:288 - [TRELLIS/API] Frame path found: /tmp/flux_view_10ccuwew/frame_5406073f-5c42-4e9a-b03d-f136dbd14931.jpg
2025-11-13 19:46:08.017 | INFO | __main__:generate_3d:292 - [TRELLIS/API] Assets found: GLB=/tmp/flux_view_10ccuwew/mesh_5406073f-5c42-4e9a-b03d-f136dbd14931.glb, PLY=/tmp/flux_view_10ccuwew/mesh_5406073f-5c42-4e9a-b03d-f136dbd14931.ply, MP4=/tmp/flux_view_10ccuwew/preview_5406073f-5c42-4e9a-b03d-f136dbd14931.mp4
2025-11-13 19:55:00.449 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: river between trees
2025-11-13 19:55:02.969 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: flux
2025-11-13 19:55:03.182 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 10
2025-11-13 19:55:03.536 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: river between trees
2025-11-13 19:55:03.536 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: river between trees (using qwen2.5:7b)
2025-11-13 19:55:04.339 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'river, between trees, full view', 'negative': 'blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background'}
2025-11-13 19:55:04.551 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-13 19:55:04.551 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-13 19:55:04.551 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-13 19:55:09.518 | INFO | __main__:_api_t2i_generate_flux:166 - [FLUX] API call successful, decoding 1 image...
2025-11-13 19:55:09.543 | INFO | __main__:_api_t2i_generate_flux:178 - Saved single FLUX image to /tmp/flux_view_3_4i0uzb for feedback
2025-11-13 19:55:09.544 | INFO | __main__:generate_image_flux:186 - [FLUX] Got 1 image
2025-11-13 19:55:09.994 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: river, between trees, full view, negative: blurry, low detail, artifacts, bad proportions, deformed parts, floating objects, shadows, lighting, background, eval:
2025-11-13 19:55:09.998 | INFO | __main__:route_post_processing:368 - [Gradio] Running FLUX post-processing (unlock only)
2025-11-13 19:55:09.998 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 19:55:10.004 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 19:55:11.361 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: river, between trees, full view, negative: bad proportions, deformed parts, artifacts, low detail, shadows, lighting, background, floating objects, eval:
2025-11-13 19:55:11.361 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 19:55:11.362 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 19:56:18.363 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/flux_view_3_4i0uzb/trellis_output.zip
2025-11-13 19:56:18.364 | INFO | __main__:generate_3d:243 - [Gradio/Fix] Cleaning old mesh files from /tmp/flux_view_3_4i0uzb
2025-11-13 19:56:18.450 | INFO | __main__:generate_3d:261 - [TRELLIS/API] Extracted ZIP to /tmp/flux_view_3_4i0uzb
2025-11-13 19:56:18.450 | INFO | __main__:generate_3d:264 - [TRELLIS/API-DEBUG] Files in /tmp/flux_view_3_4i0uzb: ['view_00.jpg', 'trellis_output.zip', 'mesh_d29b21b9-c167-41f5-92a5-f6ab26d4e72d.glb', 'mesh_d29b21b9-c167-41f5-92a5-f6ab26d4e72d.ply', 'preview_d29b21b9-c167-41f5-92a5-f6ab26d4e72d.mp4', 'frame_d29b21b9-c167-41f5-92a5-f6ab26d4e72d.jpg']
2025-11-13 19:56:18.450 | INFO | __main__:generate_3d:271 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_d29b21b9-c167-41f5-92a5-f6ab26d4e72d.glb']
2025-11-13 19:56:18.450 | INFO | __main__:generate_3d:272 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_d29b21b9-c167-41f5-92a5-f6ab26d4e72d.ply']
2025-11-13 19:56:18.450 | INFO | __main__:generate_3d:273 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_d29b21b9-c167-41f5-92a5-f6ab26d4e72d.mp4']
2025-11-13 19:56:18.450 | INFO | __main__:generate_3d:274 - [TRELLIS/API-DEBUG] Found frame files: ['frame_d29b21b9-c167-41f5-92a5-f6ab26d4e72d.jpg']
2025-11-13 19:56:18.451 | INFO | __main__:generate_3d:288 - [TRELLIS/API] Frame path found: /tmp/flux_view_3_4i0uzb/frame_d29b21b9-c167-41f5-92a5-f6ab26d4e72d.jpg
2025-11-13 19:56:18.451 | INFO | __main__:generate_3d:292 - [TRELLIS/API] Assets found: GLB=/tmp/flux_view_3_4i0uzb/mesh_d29b21b9-c167-41f5-92a5-f6ab26d4e72d.glb, PLY=/tmp/flux_view_3_4i0uzb/mesh_d29b21b9-c167-41f5-92a5-f6ab26d4e72d.ply, MP4=/tmp/flux_view_3_4i0uzb/preview_d29b21b9-c167-41f5-92a5-f6ab26d4e72d.mp4
2025-11-13 19:56:52.481 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: dog knight
2025-11-13 19:56:52.770 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: sd
2025-11-13 19:56:52.986 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 28
2025-11-13 19:56:53.289 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: dog knight
2025-11-13 19:56:53.289 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: dog knight (using qwen2.5:7b)
2025-11-13 19:56:54.065 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'dog, knight, full body', 'negative': 'blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background'}
2025-11-13 19:56:54.302 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: sd with 28 steps
2025-11-13 19:56:54.303 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 28 inference steps...
2025-11-13 19:56:54.303 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 28 steps...
2025-11-13 19:57:06.205 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 19:57:06.260 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_hr4z_uqx for Trellis step
2025-11-13 19:57:06.260 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 19:57:06.698 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: dog, knight, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval:
2025-11-13 19:57:06.698 | INFO | __main__:route_post_processing:352 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-13 19:57:06.699 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 19:57:06.700 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 19:57:06.700 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-13 19:57:08.170 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: dog, knight, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, unnatural shadows, poor lighting, cluttered background, eval:
2025-11-13 19:57:08.170 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 19:57:08.171 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 19:57:12.527 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'The images match the prompt well, showing a dog knight from different angles. The images are consistent across the views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 19:57:12.528 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'The images match the prompt well, showing a dog knight from different angles. The images are consistent across the views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 19:57:12.528 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: dog, knight, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval: The images match the prompt well, showing a dog knight from different angles. The images are consistent across the views, and there are no visible issues or artifacts.
2025-11-13 19:57:12.528 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 19:57:12.528 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 19:58:03.155 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/views_hr4z_uqx/trellis_output.zip
2025-11-13 19:58:03.156 | INFO | __main__:generate_3d:243 - [Gradio/Fix] Cleaning old mesh files from /tmp/views_hr4z_uqx
2025-11-13 19:58:03.238 | INFO | __main__:generate_3d:261 - [TRELLIS/API] Extracted ZIP to /tmp/views_hr4z_uqx
2025-11-13 19:58:03.239 | INFO | __main__:generate_3d:264 - [TRELLIS/API-DEBUG] Files in /tmp/views_hr4z_uqx: ['view_00.jpg', 'view_01.jpg', 'view_02.jpg', 'view_03.jpg', 'view_04.jpg', 'view_05.jpg', 'trellis_output.zip', 'mesh_cc212b96-1e1f-4073-8fcf-90b5567eb98b.glb', 'mesh_cc212b96-1e1f-4073-8fcf-90b5567eb98b.ply', 'preview_cc212b96-1e1f-4073-8fcf-90b5567eb98b.mp4', 'frame_cc212b96-1e1f-4073-8fcf-90b5567eb98b.jpg']
2025-11-13 19:58:03.239 | INFO | __main__:generate_3d:271 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_cc212b96-1e1f-4073-8fcf-90b5567eb98b.glb']
2025-11-13 19:58:03.239 | INFO | __main__:generate_3d:272 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_cc212b96-1e1f-4073-8fcf-90b5567eb98b.ply']
2025-11-13 19:58:03.239 | INFO | __main__:generate_3d:273 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_cc212b96-1e1f-4073-8fcf-90b5567eb98b.mp4']
2025-11-13 19:58:03.239 | INFO | __main__:generate_3d:274 - [TRELLIS/API-DEBUG] Found frame files: ['frame_cc212b96-1e1f-4073-8fcf-90b5567eb98b.jpg']
2025-11-13 19:58:03.239 | INFO | __main__:generate_3d:288 - [TRELLIS/API] Frame path found: /tmp/views_hr4z_uqx/frame_cc212b96-1e1f-4073-8fcf-90b5567eb98b.jpg
2025-11-13 19:58:03.239 | INFO | __main__:generate_3d:292 - [TRELLIS/API] Assets found: GLB=/tmp/views_hr4z_uqx/mesh_cc212b96-1e1f-4073-8fcf-90b5567eb98b.glb, PLY=/tmp/views_hr4z_uqx/mesh_cc212b96-1e1f-4073-8fcf-90b5567eb98b.ply, MP4=/tmp/views_hr4z_uqx/preview_cc212b96-1e1f-4073-8fcf-90b5567eb98b.mp4
2025-11-13 19:59:19.538 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: dog knight
2025-11-13 19:59:19.892 | INFO | __main__:call_choose_model:314 - [Gradio] LLM chose model: sd
2025-11-13 19:59:20.096 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 28
2025-11-13 19:59:20.399 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: dog knight
2025-11-13 19:59:20.400 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: dog knight (using qwen2.5:7b)
2025-11-13 19:59:21.188 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'dog, knight, full body', 'negative': 'blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background'}
2025-11-13 19:59:21.419 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: sd with 28 steps
2025-11-13 19:59:21.419 | INFO | __main__:generate_images:112 - [MV-ADAPTER] Starting generation with 28 inference steps...
2025-11-13 19:59:21.420 | INFO | __main__:_api_t2mv_generate:77 - [MV-ADAPTER] Calling API with 28 steps...
2025-11-13 19:59:33.329 | INFO | __main__:_api_t2mv_generate:94 - [MV-ADAPTER] API call successful, decoding 6 images...
2025-11-13 19:59:33.381 | INFO | __main__:_api_t2mv_generate:107 - Saved images to /tmp/views_t9khy0_7 for Trellis step
2025-11-13 19:59:33.382 | INFO | __main__:generate_images:116 - [MV-ADAPTER] Got 6 images
2025-11-13 19:59:33.765 | INFO | __main__:route_post_processing:352 - [Gradio] Running SD post-processing (check_views, chat_start)
2025-11-13 19:59:33.766 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: dog, knight, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval:
2025-11-13 19:59:33.767 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 19:59:33.767 | INFO | ollama:check_views:228 - [MV-ADAPTER] check_views loading (using qwen2.5vl:7b)
2025-11-13 19:59:33.768 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 19:59:34.974 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: dog, full body knight, negative: low detail, artifacts, bad anatomy, extra limbs, deformed face, shadows, lighting, background, eval:
2025-11-13 19:59:34.974 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 19:59:34.975 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 19:59:39.260 | INFO | ollama:check_views:235 - [MV-ADAPTER] check_views result: {'avaliation': 'The images match the prompt well, showing a dog knight from different angles. The images are consistent across the views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 19:59:39.261 | INFO | __main__:check_views_quality:145 - [MV-ADAPTER] views quality check: {'avaliation': 'The images match the prompt well, showing a dog knight from different angles. The images are consistent across the views, and there are no visible issues or artifacts.', 'negative_prompt': 'none'}
2025-11-13 19:59:39.261 | INFO | __main__:chat_start_fn:35 - [CHAT] starting chat with refined: dog, knight, full body, negative: blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background, eval: The images match the prompt well, showing a dog knight from different angles. The images are consistent across the views, and there are no visible issues or artifacts.
2025-11-13 19:59:39.261 | INFO | ollama_chat:start_chat_from_inputs:112 - [CHAT] STARTING CHAT (using qwen2.5:7b)
2025-11-13 19:59:39.261 | INFO | ollama_chat:chat_call:97 - [CHAT] chat_call using qwen2.5:7b
2025-11-13 20:00:36.225 | INFO | __main__:generate_3d:234 - [TRELLIS/API] Saved ZIP to /tmp/views_t9khy0_7/trellis_output.zip
2025-11-13 20:00:36.226 | INFO | __main__:generate_3d:243 - [Gradio/Fix] Cleaning old mesh files from /tmp/views_t9khy0_7
2025-11-13 20:00:36.317 | INFO | __main__:generate_3d:261 - [TRELLIS/API] Extracted ZIP to /tmp/views_t9khy0_7
2025-11-13 20:00:36.317 | INFO | __main__:generate_3d:264 - [TRELLIS/API-DEBUG] Files in /tmp/views_t9khy0_7: ['view_00.jpg', 'view_01.jpg', 'view_02.jpg', 'view_03.jpg', 'view_04.jpg', 'view_05.jpg', 'trellis_output.zip', 'mesh_0983b189-a8f4-4165-9e56-0f97949f1589.glb', 'mesh_0983b189-a8f4-4165-9e56-0f97949f1589.ply', 'preview_0983b189-a8f4-4165-9e56-0f97949f1589.mp4', 'frame_0983b189-a8f4-4165-9e56-0f97949f1589.jpg']
2025-11-13 20:00:36.318 | INFO | __main__:generate_3d:271 - [TRELLIS/API-DEBUG] Found GLB files: ['mesh_0983b189-a8f4-4165-9e56-0f97949f1589.glb']
2025-11-13 20:00:36.318 | INFO | __main__:generate_3d:272 - [TRELLIS/API-DEBUG] Found PLY files: ['mesh_0983b189-a8f4-4165-9e56-0f97949f1589.ply']
2025-11-13 20:00:36.318 | INFO | __main__:generate_3d:273 - [TRELLIS/API-DEBUG] Found MP4 files: ['preview_0983b189-a8f4-4165-9e56-0f97949f1589.mp4']
2025-11-13 20:00:36.318 | INFO | __main__:generate_3d:274 - [TRELLIS/API-DEBUG] Found frame files: ['frame_0983b189-a8f4-4165-9e56-0f97949f1589.jpg']
2025-11-13 20:00:36.318 | INFO | __main__:generate_3d:288 - [TRELLIS/API] Frame path found: /tmp/views_t9khy0_7/frame_0983b189-a8f4-4165-9e56-0f97949f1589.jpg
2025-11-13 20:00:36.318 | INFO | __main__:generate_3d:292 - [TRELLIS/API] Assets found: GLB=/tmp/views_t9khy0_7/mesh_0983b189-a8f4-4165-9e56-0f97949f1589.glb, PLY=/tmp/views_t9khy0_7/mesh_0983b189-a8f4-4165-9e56-0f97949f1589.ply, MP4=/tmp/views_t9khy0_7/preview_0983b189-a8f4-4165-9e56-0f97949f1589.mp4
2025-11-14 11:49:18.870 | INFO | __main__:call_choose_model:309 - [Gradio] Choosing model for prompt: dog knight
2025-11-14 11:49:18.873 | ERROR | __main__:call_choose_model:317 - [Gradio] Failed to choose model: HTTPConnectionPool(host='127.0.0.1', port=8081): Max retries exceeded with url: /t2g/choose_model (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7ff33ed5aaa0>: Failed to establish a new connection: [Errno 111] Connection refused')). Defaulting to 'flux'.
2025-11-14 11:49:19.223 | INFO | __main__:_reset_ui_and_set_steps:323 - [Gradio] Resetting UI, setting steps to 10
2025-11-14 11:49:19.526 | INFO | __main__:llm_prompt_processing:65 - [MV-ADAPTER] generating images. prompt to be refined: dog knight
2025-11-14 11:49:19.526 | INFO | ollama:prepare_prompts:133 - [OLLAMA] prepare_prompts: preparing prompts for input: dog knight (using qwen2.5:7b)
2025-11-14 11:49:22.797 | INFO | ollama:prepare_prompts:169 - [MV-ADAPTER] prepare_prompts result: {'refined': 'dog, knight, full body', 'negative': 'blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background'}
2025-11-14 11:49:23.011 | INFO | __main__:route_generation:343 - [Gradio] Routing generation to model: flux with 10 steps
2025-11-14 11:49:23.011 | INFO | __main__:generate_image_flux:182 - [FLUX] Starting generation with 10 inference steps...
2025-11-14 11:49:23.012 | INFO | __main__:_api_t2i_generate_flux:151 - [FLUX] Calling API with 10 steps...
2025-11-14 11:49:23.014 | ERROR | __main__:generate_image_flux:198 - [FLUX] Error generating image: HTTPConnectionPool(host='127.0.0.1', port=8081): Max retries exceeded with url: /t2i/generate_flux?ref_prompt=dog%2C+knight%2C+full+body&neg_prompt=blurry%2C+low+detail%2C+artifacts%2C+bad+anatomy%2C+extra+limbs%2C+deformed+face%2C+deformed+tail%2C+shadows%2C+lighting%2C+background&randomize=False&inference_steps=10 (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7ff33edc8bb0>: Failed to establish a new connection: [Errno 111] Connection refused'))
Traceback (most recent call last):
File "/venv/beau/lib/python3.10/site-packages/urllib3/connection.py", line 198, in _new_conn
sock = connection.create_connection(
│ └ <function create_connection at 0x7ff4a7d8d090>
└ <module 'urllib3.util.connection' from '/venv/beau/lib/python3.10/site-packages/urllib3/util/connection.py'>
File "/venv/beau/lib/python3.10/site-packages/urllib3/util/connection.py", line 85, in create_connection
raise err
└ None
File "/venv/beau/lib/python3.10/site-packages/urllib3/util/connection.py", line 73, in create_connection
sock.connect(sa)
│ │ └ ('127.0.0.1', 8081)
│ └ <method 'connect' of '_socket.socket' objects>
└ <socket.socket [closed] fd=-1, family=AddressFamily.AF_INET, type=SocketKind.SOCK_STREAM, proto=6>
ConnectionRefusedError: [Errno 111] Connection refused
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/venv/beau/lib/python3.10/site-packages/urllib3/connectionpool.py", line 787, in urlopen
response = self._make_request(
│ └ <function HTTPConnectionPool._make_request at 0x7ff4a79cdf30>
└ <urllib3.connectionpool.HTTPConnectionPool object at 0x7ff32fb23d90>
File "/venv/beau/lib/python3.10/site-packages/urllib3/connectionpool.py", line 493, in _make_request
conn.request(
│ └ <function HTTPConnection.request at 0x7ff4a79ba440>
└ <urllib3.connection.HTTPConnection object at 0x7ff33edc8bb0>
File "/venv/beau/lib/python3.10/site-packages/urllib3/connection.py", line 494, in request
self.endheaders()
│ └ <function HTTPConnection.endheaders at 0x7ff4a8676f80>
└ <urllib3.connection.HTTPConnection object at 0x7ff33edc8bb0>
File "/venv/beau/lib/python3.10/http/client.py", line 1278, in endheaders
self._send_output(message_body, encode_chunked=encode_chunked)
│ │ │ └ False
│ │ └ None
│ └ <function HTTPConnection._send_output at 0x7ff4a8676b90>
└ <urllib3.connection.HTTPConnection object at 0x7ff33edc8bb0>
File "/venv/beau/lib/python3.10/http/client.py", line 1038, in _send_output
self.send(msg)
│ │ └ b'POST /t2i/generate_flux?ref_prompt=dog%2C+knight%2C+full+body&neg_prompt=blurry%2C+low+detail%2C+artifacts%2C+bad+anatomy%2...
│ └ <function HTTPConnection.send at 0x7ff4a86769e0>
└ <urllib3.connection.HTTPConnection object at 0x7ff33edc8bb0>
File "/venv/beau/lib/python3.10/http/client.py", line 976, in send
self.connect()
│ └ <function HTTPConnection.connect at 0x7ff4a79b9f30>
└ <urllib3.connection.HTTPConnection object at 0x7ff33edc8bb0>
File "/venv/beau/lib/python3.10/site-packages/urllib3/connection.py", line 325, in connect
self.sock = self._new_conn()
│ │ │ └ <function HTTPConnection._new_conn at 0x7ff4a79b9cf0>
│ │ └ <urllib3.connection.HTTPConnection object at 0x7ff33edc8bb0>
│ └ None
└ <urllib3.connection.HTTPConnection object at 0x7ff33edc8bb0>
File "/venv/beau/lib/python3.10/site-packages/urllib3/connection.py", line 213, in _new_conn
raise NewConnectionError(
└ <class 'urllib3.exceptions.NewConnectionError'>
urllib3.exceptions.NewConnectionError: <urllib3.connection.HTTPConnection object at 0x7ff33edc8bb0>: Failed to establish a new connection: [Errno 111] Connection refused
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/venv/beau/lib/python3.10/site-packages/requests/adapters.py", line 644, in send
resp = conn.urlopen(
│ └ <function HTTPConnectionPool.urlopen at 0x7ff4a79ce0e0>
└ <urllib3.connectionpool.HTTPConnectionPool object at 0x7ff32fb23d90>
File "/venv/beau/lib/python3.10/site-packages/urllib3/connectionpool.py", line 841, in urlopen
retries = retries.increment(
│ └ <function Retry.increment at 0x7ff4a85c9630>
└ Retry(total=0, connect=None, read=False, redirect=None, status=None)
File "/venv/beau/lib/python3.10/site-packages/urllib3/util/retry.py", line 519, in increment
raise MaxRetryError(_pool, url, reason) from reason # type: ignore[arg-type]
│ │ │ │ └ NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7ff33edc8bb0>: Failed to establish a new connection: [Errn...
│ │ │ └ NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7ff33edc8bb0>: Failed to establish a new connection: [Errn...
│ │ └ '/t2i/generate_flux?ref_prompt=dog%2C+knight%2C+full+body&neg_prompt=blurry%2C+low+detail%2C+artifacts%2C+bad+anatomy%2C+extr...
│ └ <urllib3.connectionpool.HTTPConnectionPool object at 0x7ff32fb23d90>
└ <class 'urllib3.exceptions.MaxRetryError'>
urllib3.exceptions.MaxRetryError: HTTPConnectionPool(host='127.0.0.1', port=8081): Max retries exceeded with url: /t2i/generate_flux?ref_prompt=dog%2C+knight%2C+full+body&neg_prompt=blurry%2C+low+detail%2C+artifacts%2C+bad+anatomy%2C+extra+limbs%2C+deformed+face%2C+deformed+tail%2C+shadows%2C+lighting%2C+background&randomize=False&inference_steps=10 (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7ff33edc8bb0>: Failed to establish a new connection: [Errno 111] Connection refused'))
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/venv/beau/lib/python3.10/threading.py", line 973, in _bootstrap
self._bootstrap_inner()
│ └ <function Thread._bootstrap_inner at 0x7ff4a874f010>
└ <WorkerThread(AnyIO worker thread, started daemon 140682444871360)>
File "/venv/beau/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
self.run()
│ └ <function WorkerThread.run at 0x7ff34d09b1c0>
└ <WorkerThread(AnyIO worker thread, started daemon 140682444871360)>
File "/venv/beau/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 976, in run
result = context.run(func, *args)
│ │ │ └ ('flux', 'dog, knight, full body', 'blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, sh...
│ │ └ <function route_generation at 0x7ff33edcc790>
│ └ <method 'run' of '_contextvars.Context' objects>
└ <_contextvars.Context object at 0x7ff33ed4eb00>
File "/venv/beau/lib/python3.10/site-packages/gradio/utils.py", line 915, in wrapper
response = f(*args, **kwargs)
│ │ └ {}
│ └ ('flux', 'dog, knight, full body', 'blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, sh...
└ <function route_generation at 0x7ff35261a4d0>
File "/workspace/beau/beau-pipeline/demo_chat_apiv3.py", line 347, in route_generation
return generate_image_flux(refined_prompt, negative_prompt, randomize, inference_steps)
│ │ │ │ └ 10
│ │ │ └ False
│ │ └ 'blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background'
│ └ 'dog, knight, full body'
└ <function generate_image_flux at 0x7ff35261a200>
> File "/workspace/beau/beau-pipeline/demo_chat_apiv3.py", line 184, in generate_image_flux
pil_images, views_dir = _api_t2i_generate_flux(refined_prompt, negative_prompt, randomize, inference_steps)
│ │ │ │ └ 10
│ │ │ └ False
│ │ └ 'blurry, low detail, artifacts, bad anatomy, extra limbs, deformed face, deformed tail, shadows, lighting, background'
│ └ 'dog, knight, full body'
└ <function _api_t2i_generate_flux at 0x7ff35261a170>