工作流 /
视频 /
混元 视频 图像 转 视频 v2（混元 视频）
混元 视频 图像 转 视频 v2（混元 视频）
hunyuan video image to video v2
ComfyUI 官方示例：混元 视频 功能演示，可拖进 ComfyUI 直接使用。
视频
22 个节点 Comfy-Org 17 KB
需要的权重
这个工作流引用了 5 个权重文件，其中 5 个能在本站模型库里找到。
clip_l.safetensors
在模型库查看
hunyuan_video_v2_replace_image_to_video_720p_bf16.safetensors
在模型库查看
hunyuan_video_vae_bf16.safetensors
在模型库查看
llava_llama3_fp8_scaled.safetensors
在模型库查看
llava_llama3_vision.safetensors
在模型库查看
节点构成
Note
× 3
BasicGuider
× 1
VAEDecode
× 1
SamplerCustomAdvanced
× 1
FluxGuidance
× 1
ModelSamplingSD3
× 1
BasicScheduler
× 1
KSamplerSelect
× 1
CLIPVisionEncode
× 1
LoadImage
× 1
CLIPVisionLoader
× 1
VAELoader
× 1
DualCLIPLoader
× 1
HunyuanImageToVideo
× 1
VAEDecodeTiled
× 1
SaveAnimatedWEBP
× 1
官方标注的模型
clip_l.safetensorshunyuan_video_v2_replace_image_to_video_720p_bf16.safetensorshunyuan_video_vae_bf16.safetensorsllava_llama3_fp8_scaled.safetensorsllava_llama3_vision.safetensors
工作流 JSON
和上面「下载工作流 JSON」是同一份文件。
1{
2 "last_node_id": 90,
3 "last_link_id": 233,
4 "nodes": [
5 {
6 "id": 22,
7 "type": "BasicGuider",
8 "pos": [
9 600,
10 0
11 ],
12 "size": [
13 222.3482666015625,
14 46
15 ],
16 "flags": [],
17 "order": 16,
18 "mode": 0,
19 "inputs": [
20 {
21 "name": "model",
22 "type": "MODEL",
23 "link": 195,
24 "slot_index": 0
25 },
26 {
27 "name": "conditioning",
28 "type": "CONDITIONING",
29 "link": 129,
30 "slot_index": 1
31 }
32 ],
33 "outputs": [
34 {
35 "name": "GUIDER",
36 "type": "GUIDER",
37 "shape": 3,
38 "links": [
39 30
40 ],
41 "slot_index": 0
42 }
43 ],
44 "properties": {
45 "Node name for S&R": "BasicGuider"
46 },
47 "widgets_values": []
48 },
49 {
50 "id": 8,
51 "type": "VAEDecode",
52 "pos": [
53 1150,
54 90
55 ],
56 "size": [
57 210,
58 46
59 ],
60 "flags": [],
61 "order": 18,
62 "mode": 2,
63 "inputs": [
64 {
65 "name": "samples",
66 "type": "LATENT",
67 "link": 181
68 },
69 {
70 "name": "vae",
71 "type": "VAE",
72 "link": 206
73 }
74 ],
75 "outputs": [
76 {
77 "name": "IMAGE",
78 "type": "IMAGE",
79 "links": [],
80 "slot_index": 0
81 }
82 ],
83 "properties": {
84 "Node name for S&R": "VAEDecode"
85 },
86 "widgets_values": []
87 },
88 {
89 "id": 13,
90 "type": "SamplerCustomAdvanced",
91 "pos": [
92 860,
93 200
94 ],
95 "size": [
96 272.3617858886719,
97 124.53733825683594
98 ],
99 "flags": [],
100 "order": 17,
101 "mode": 0,
102 "inputs": [
103 {
104 "name": "noise",
105 "type": "NOISE",
106 "link": 37,
107 "slot_index": 0
108 },
109 {
110 "name": "guider",
111 "type": "GUIDER",
112 "link": 30,
113 "slot_index": 1
114 },
115 {
116 "name": "sampler",
117 "type": "SAMPLER",
118 "link": 19,
119 "slot_index": 2
120 },
121 {
122 "name": "sigmas",
123 "type": "SIGMAS",
124 "link": 20,
125 "slot_index": 3
126 },
127 {
128 "name": "latent_image",
129 "type": "LATENT",
130 "link": 216,
131 "slot_index": 4
132 }
133 ],
134 "outputs": [
135 {
136 "name": "output",
137 "type": "LATENT",
138 "shape": 3,
139 "links": [
140 181,
141 210
142 ],
143 "slot_index": 0
144 },
145 {
146 "name": "denoised_output",
147 "type": "LATENT",
148 "shape": 3,
149 "links": null
150 }
151 ],
152 "properties": {
153 "Node name for S&R": "SamplerCustomAdvanced"
154 },
155 "widgets_values": []
156 },
157 {
158 "id": 74,
159 "type": "Note",
160 "pos": [
161 1147.7459716796875,
162 405.0789489746094
163 ],
164 "size": [
165 210,
166 170
167 ],
168 "flags": [],
169 "order": 0,
170 "mode": 0,
171 "inputs": [],
172 "outputs": [],
173 "properties": [],
174 "widgets_values": [
175 "Use the tiled decode node by default because most people will need it.\n\nLower the tile_size and overlap if you run out of memory."
176 ],
177 "color": "#432",
178 "bgcolor": "#653"
179 },
180 {
181 "id": 26,
182 "type": "FluxGuidance",
183 "pos": [
184 514.2149047851562,
185 86.77685546875
186 ],
187 "size": [
188 317.4000244140625,
189 58
190 ],
191 "flags": [],
192 "order": 15,
193 "mode": 0,
194 "inputs": [
195 {
196 "name": "conditioning",
197 "type": "CONDITIONING",
198 "link": 225
199 }
200 ],
201 "outputs": [
202 {
203 "name": "CONDITIONING",
204 "type": "CONDITIONING",
205 "shape": 3,
206 "links": [
207 129
208 ],
209 "slot_index": 0
210 }
211 ],
212 "properties": {
213 "Node name for S&R": "FluxGuidance"
214 },
215 "widgets_values": [
216 6
217 ],
218 "color": "#233",
219 "bgcolor": "#355"
220 },
221 {
222 "id": 67,
223 "type": "ModelSamplingSD3",
224 "pos": [
225 360,
226 0
227 ],
228 "size": [
229 210,
230 58
231 ],
232 "flags": [],
233 "order": 12,
234 "mode": 0,
235 "inputs": [
236 {
237 "name": "model",
238 "type": "MODEL",
239 "link": 209
240 }
241 ],
242 "outputs": [
243 {
244 "name": "MODEL",
245 "type": "MODEL",
246 "links": [
247 195
248 ],
249 "slot_index": 0
250 }
251 ],
252 "properties": {
253 "Node name for S&R": "ModelSamplingSD3"
254 },
255 "widgets_values": [
256 7
257 ]
258 },
259 {
260 "id": 17,
261 "type": "BasicScheduler",
262 "pos": [
263 510,
264 660
265 ],
266 "size": [
267 315,
268 106
269 ],
270 "flags": [],
271 "order": 11,
272 "mode": 0,
273 "inputs": [
274 {
275 "name": "model",
276 "type": "MODEL",
277 "link": 190,
278 "slot_index": 0
279 }
280 ],
281 "outputs": [
282 {
283 "name": "SIGMAS",
284 "type": "SIGMAS",
285 "shape": 3,
286 "links": [
287 20
288 ]
289 }
290 ],
291 "properties": {
292 "Node name for S&R": "BasicScheduler"
293 },
294 "widgets_values": [
295 "simple",
296 20,
297 1
298 ]
299 },
300 {
301 "id": 16,
302 "type": "KSamplerSelect",
303 "pos": [
304 520,
305 550
306 ],
307 "size": [
308 315,
309 58
310 ],
311 "flags": [],
312 "order": 1,
313 "mode": 0,
314 "inputs": [],
315 "outputs": [
316 {
317 "name": "SAMPLER",
318 "type": "SAMPLER",
319 "shape": 3,
320 "links": [
321 19
322 ]
323 }
324 ],
325 "properties": {
326 "Node name for S&R": "KSamplerSelect"
327 },
328 "widgets_values": [
329 "euler"
330 ]
331 },
332 {
333 "id": 81,
334 "type": "CLIPVisionEncode",
335 "pos": [
336 200,
337 530
338 ],
339 "size": [
340 253.60000610351562,
341 78
342 ],
343 "flags": [],
344 "order": 10,
345 "mode": 0,
346 "inputs": [
347 {
348 "name": "clip_vision",
349 "type": "CLIP_VISION",
350 "link": 220
351 },
352 {
353 "name": "image",
354 "type": "IMAGE",
355 "link": 221
356 }
357 ],
358 "outputs": [
359 {
360 "name": "CLIP_VISION_OUTPUT",
361 "type": "CLIP_VISION_OUTPUT",
362 "links": [
363 219
364 ]
365 }
366 ],
367 "properties": {
368 "Node name for S&R": "CLIPVisionEncode"
369 },
370 "widgets_values": [
371 "none"
372 ]
373 },
374 {
375 "id": 89,
376 "type": "Note",
377 "pos": [
378 190,
379 660
380 ],
381 "size": [
382 260,
383 210
384 ],
385 "flags": [],
386 "order": 2,
387 "mode": 0,
388 "inputs": [],
389 "outputs": [],
390 "properties": [],
391 "widgets_values": [
392 "Select your image here. The image is fed to both the text encoder and directly to the model.\n\nYou can set the resolution and length of the video using the HunyuanImageToVideo node."
393 ],
394 "color": "#432",
395 "bgcolor": "#653"
396 },
397 {
398 "id": 83,
399 "type": "LoadImage",
400 "pos": [
401 -190,
402 700
403 ],
404 "size": [
405 365.4132080078125,
406 471.8512268066406
407 ],
408 "flags": [],
409 "order": 3,
410 "mode": 0,
411 "inputs": [],
412 "outputs": [
413 {
414 "name": "IMAGE",
415 "type": "IMAGE",
416 "links": [
417 221,
418 222
419 ],
420 "slot_index": 0
421 },
422 {
423 "name": "MASK",
424 "type": "MASK",
425 "links": null
426 }
427 ],
428 "properties": {
429 "Node name for S&R": "LoadImage"
430 },
431 "widgets_values": [
432 "flux_dev_example.png",
433 "image"
434 ]
435 },
436 {
437 "id": 82,
438 "type": "CLIPVisionLoader",
439 "pos": [
440 -190,
441 580
442 ],
443 "size": [
444 315,
445 58
446 ],
447 "flags": [],
448 "order": 4,
449 "mode": 0,
450 "inputs": [],
451 "outputs": [
452 {
453 "name": "CLIP_VISION",
454 "type": "CLIP_VISION",
455 "links": [
456 220
457 ]
458 }
459 ],
460 "properties": {
461 "Node name for S&R": "CLIPVisionLoader"
462 },
463 "widgets_values": [
464 "llava_llama3_vision.safetensors"
465 ],
466 "color": "#223",
467 "bgcolor": "#335"
468 },
469 {
470 "id": 10,
471 "type": "VAELoader",
472 "pos": [
473 -190,
474 470
475 ],
476 "size": [
477 350,
478 60
479 ],
480 "flags": [],
481 "order": 5,
482 "mode": 0,
483 "inputs": [],
484 "outputs": [
485 {
486 "name": "VAE",
487 "type": "VAE",
488 "shape": 3,
489 "links": [
490 206,
491 211,
492 223
493 ],
494 "slot_index": 0
495 }
496 ],
497 "properties": {
498 "Node name for S&R": "VAELoader"
499 },
500 "widgets_values": [
501 "hunyuan_video_vae_bf16.safetensors"
502 ],
503 "color": "#223",
504 "bgcolor": "#335"
505 },
506 {
507 "id": 11,
508 "type": "DualCLIPLoader",
509 "pos": [
510 -190,
511 290
512 ],
513 "size": [
514 350,
515 122
516 ],
517 "flags": [],
518 "order": 6,
519 "mode": 0,
520 "inputs": [],
521 "outputs": [
522 {
523 "name": "CLIP",
524 "type": "CLIP",
525 "shape": 3,
526 "links": [
527 224
528 ],
529 "slot_index": 0
530 }
531 ],
532 "properties": {
533 "Node name for S&R": "DualCLIPLoader"
534 },
535 "widgets_values": [
536 "clip_l.safetensors",
537 "llava_llama3_fp8_scaled.safetensors",
538 "hunyuan_video",
539 "default"
540 ],
541 "color": "#223",
542 "bgcolor": "#335"
543 },
544 {
545 "id": 77,
546 "type": "Note",
547 "pos": [
548 -140,
549 0
550 ],
551 "size": [
552 350,
553 110
554 ],
555 "flags": [],
556 "order": 7,
557 "mode": 0,
558 "inputs": [],
559 "outputs": [],
560 "properties": [],
561 "widgets_values": [
562 "Select a fp8 weight_dtype if you are running out of memory."
563 ],
564 "color": "#432",
565 "bgcolor": "#653"
566 },
567 {
568 "id": 78,
569 "type": "HunyuanImageToVideo",
570 "pos": [
571 510,
572 820
573 ],
574 "size": [
575 315,
576 194
577 ],
578 "flags": [],
579 "order": 14,
580 "mode": 0,
581 "inputs": [
582 {
583 "name": "positive",
584 "type": "CONDITIONING",
585 "link": 218
586 },
587 {
588 "name": "vae",
589 "type": "VAE",
590 "link": 223
591 },
592 {
593 "name": "start_image",
594 "type": "IMAGE",
595 "shape": 7,
596 "link": 222
597 }
598 ],
599 "outputs": [
600 {
601 "name": "positive",
602 "type": "CONDITIONING",
603 "links": [
604 225
605 ],
606 "slot_index": 0
607 },
608 {
609 "name": "latent",
610 "type": "LATENT",
611 "links": [
612 216
613 ]
614 }
615 ],
616 "properties": {
617 "Node name for S&R": "HunyuanImageToVideo"
618 },
619 "widgets_values": [
620 720,
621 720,
622 57,
623 1,
624 "v2 (replace)"
625 ]
626 },
627 {
628 "id": 73,
629 "type": "VAEDecodeTiled",
630 "pos": [
631 1150,
632 200
633 ],
634 "size": [
635 210,
636 150
637 ],
638 "flags": [],
639 "order": 19,
640 "mode": 0,
641 "inputs": [
642 {
643 "name": "samples",
644 "type": "LATENT",
645 "link": 210
646 },
647 {
648 "name": "vae",
649 "type": "VAE",
650 "link": 211
651 }
652 ],
653 "outputs": [
654 {
655 "name": "IMAGE",
656 "type": "IMAGE",
657 "links": [
658 232,
659 233
660 ],
661 "slot_index": 0
662 }
663 ],
664 "properties": {
665 "Node name for S&R": "VAEDecodeTiled"
666 },
667 "widgets_values": [
668 256,
669 64,
670 64,
671 8
672 ]
673 },
674 {
675 "id": 75,
676 "type": "SaveAnimatedWEBP",
677 "pos": [
678 1390,
679 200
680 ],
681 "size": [
682 621.495361328125,
683 587.12451171875
684 ],
685 "flags": [],
686 "order": 20,
687 "mode": 0,
688 "inputs": [
689 {
690 "name": "images",
691 "type": "IMAGE",
692 "link": 232
693 }
694 ],
695 "outputs": [],
696 "properties": [],
697 "widgets_values": [
698 "ComfyUI",
699 24,
700 false,
701 90,
702 "default"
703 ]
704 },
705 {
706 "id": 84,
707 "type": "SaveWEBM",
708 "pos": [
709 2030,
710 200
711 ],
712 "size": [
713 315,
714 130
715 ],
716 "flags": [],
717 "order": 21,
718 "mode": 4,
719 "inputs": [
720 {
721 "name": "images",
722 "type": "IMAGE",
723 "link": 233
724 }
725 ],
726 "outputs": [],
727 "properties": {
728 "Node name for S&R": "SaveWEBM"
729 },
730 "widgets_values": [
731 "ComfyUI",
732 "vp9",
733 24,
734 12
735 ]
736 },
737 {
738 "id": 80,
739 "type": "TextEncodeHunyuanVideo_ImageToVideo",
740 "pos": [
741 390,
742 180
743 ],
744 "size": [
745 441,
746 200
747 ],
748 "flags": [],
749 "order": 13,
750 "mode": 0,
751 "inputs": [
752 {
753 "name": "clip",
754 "type": "CLIP",
755 "link": 224
756 },
757 {
758 "name": "clip_vision_output",
759 "type": "CLIP_VISION_OUTPUT",
760 "link": 219
761 }
762 ],
763 "outputs": [
764 {
765 "name": "CONDITIONING",
766 "type": "CONDITIONING",
767 "links": [
768 218
769 ]
770 }
771 ],
772 "title": "Text Encode Hunyuan Video (ImageToVideo)",
773 "properties": {
774 "Node name for S&R": "TextEncodeHunyuanVideo_ImageToVideo"
775 },
776 "widgets_values": [
777 "a cute anime girl with massive fennec ears and a big fluffy tail wearing a maid outfit walking forward",
778 4
779 ],
780 "color": "#232",
781 "bgcolor": "#353"
782 },
783 {
784 "id": 25,
785 "type": "RandomNoise",
786 "pos": [
787 520,
788 420
789 ],
790 "size": [
791 315,
792 82
793 ],
794 "flags": [],
795 "order": 8,
796 "mode": 0,
797 "inputs": [],
798 "outputs": [
799 {
800 "name": "NOISE",
801 "type": "NOISE",
802 "shape": 3,
803 "links": [
804 37
805 ]
806 }
807 ],
808 "properties": {
809 "Node name for S&R": "RandomNoise"
810 },
811 "widgets_values": [
812 72275780843998,
813 "randomize"
814 ],
815 "color": "#2a363b",
816 "bgcolor": "#3f5159"
817 },
818 {
819 "id": 12,
820 "type": "UNETLoader",
821 "pos": [
822 -190,
823 160
824 ],
825 "size": [
826 404.6181640625,
827 82
828 ],
829 "flags": [],
830 "order": 9,
831 "mode": 0,
832 "inputs": [],
833 "outputs": [
834 {
835 "name": "MODEL",
836 "type": "MODEL",
837 "shape": 3,
838 "links": [
839 190,
840 209
841 ],
842 "slot_index": 0
843 }
844 ],
845 "properties": {
846 "Node name for S&R": "UNETLoader"
847 },
848 "widgets_values": [
849 "hunyuan_video_v2_replace_image_to_video_720p_bf16.safetensors",
850 "default"
851 ],
852 "color": "#223",
853 "bgcolor": "#335"
854 }
855 ],
856 "links": [
857 [
858 19,
859 16,
860 0,
861 13,
862 2,
863 "SAMPLER"
864 ],
865 [
866 20,
867 17,
868 0,
869 13,
870 3,
871 "SIGMAS"
872 ],
873 [
874 30,
875 22,
876 0,
877 13,
878 1,
879 "GUIDER"
880 ],
881 [
882 37,
883 25,
884 0,
885 13,
886 0,
887 "NOISE"
888 ],
889 [
890 129,
891 26,
892 0,
893 22,
894 1,
895 "CONDITIONING"
896 ],
897 [
898 181,
899 13,
900 0,
901 8,
902 0,
903 "LATENT"
904 ],
905 [
906 190,
907 12,
908 0,
909 17,
910 0,
911 "MODEL"
912 ],
913 [
914 195,
915 67,
916 0,
917 22,
918 0,
919 "MODEL"
920 ],
921 [
922 206,
923 10,
924 0,
925 8,
926 1,
927 "VAE"
928 ],
929 [
930 209,
931 12,
932 0,
933 67,
934 0,
935 "MODEL"
936 ],
937 [
938 210,
939 13,
940 0,
941 73,
942 0,
943 "LATENT"
944 ],
945 [
946 211,
947 10,
948 0,
949 73,
950 1,
951 "VAE"
952 ],
953 [
954 216,
955 78,
956 1,
957 13,
958 4,
959 "LATENT"
960 ],
961 [
962 218,
963 80,
964 0,
965 78,
966 0,
967 "CONDITIONING"
968 ],
969 [
970 219,
971 81,
972 0,
973 80,
974 1,
975 "CLIP_VISION_OUTPUT"
976 ],
977 [
978 220,
979 82,
980 0,
981 81,
982 0,
983 "CLIP_VISION"
984 ],
985 [
986 221,
987 83,
988 0,
989 81,
990 1,
991 "IMAGE"
992 ],
993 [
994 222,
995 83,
996 0,
997 78,
998 2,
999 "IMAGE"
1000 ],
1001 [
1002 223,
1003 10,
1004 0,
1005 78,
1006 1,
1007 "VAE"
1008 ],
1009 [
1010 224,
1011 11,
1012 0,
1013 80,
1014 0,
1015 "CLIP"
1016 ],
1017 [
1018 225,
1019 78,
1020 0,
1021 26,
1022 0,
1023 "CONDITIONING"
1024 ],
1025 [
1026 232,
1027 73,
1028 0,
1029 75,
1030 0,
1031 "IMAGE"
1032 ],
1033 [
1034 233,
1035 73,
1036 0,
1037 84,
1038 0,
1039 "IMAGE"
1040 ]
1041 ],
1042 "groups": [],
1043 "config": [],
1044 "extra": {
1045 "groupNodes": []
1046 },
1047 "version": 0.4
1048}