ComfyPilotComfyPilot
工作流 / 视频 / Kandinsky 5.0 Video Lite 文生视频
Kandinsky 5.0 Video Lite 文生视频

Kandinsky 5.0 Video Lite 文生视频

Kandinsky 5.0 Video Lite Text to Video

一款轻量级2B模型，可根据英文和俄文提示生成高质量视频。

视频 4 个节点 14 次使用 ComfyUI ≥ 0.4.0 ComfyUI 23 KB

下载工作流 JSON 查看官方来源

下载后把文件拖进 ComfyUI 窗口即可打开。

需要的权重

这个工作流引用了 4 个权重文件，其中 3 个能在本站模型库里找到。

clip_l.safetensors 在模型库查看
hunyuan_video_vae_bf16.safetensors 在模型库查看
kandinsky5lite_t2v_sft_5s.safetensors 模型库暂未收录
qwen_2.5_vl_7b_fp8_scaled.safetensors 在模型库查看

节点构成

MarkdownNote × 2
7aad998c-49e7-433f-bfb9-b1ac2680aa9e × 1
SaveVideo × 1

官方标注的模型

Kandinsky

工作流 JSON

和上面「下载工作流 JSON」是同一份文件。

1{
2 "id": "188a3615-701f-49aa-8765-85687e960e23",
3 "revision": 0,
4 "last_node_id": 15,
5 "last_link_id": 15,
6 "nodes": [
7 {
8 "id": 12,
9 "type": "7aad998c-49e7-433f-bfb9-b1ac2680aa9e",
10 "pos": [
11 900,
12 730
13 ],
14 "size": [
15 400,
16 470
17 ],
18 "flags": [],
19 "order": 0,
20 "mode": 0,
21 "inputs": [],
22 "outputs": [
23 {
24 "name": "VIDEO",
25 "type": "VIDEO",
26 "links": [
27 15
28 ]
29 }
30 ],
31 "properties": {
32 "proxyWidgets": [
33 [
34 "7",
35 "text"
36 ],
37 [
38 "5",
39 "width"
40 ],
41 [
42 "5",
43 "height"
44 ],
45 [
46 "5",
47 "length"
48 ],
49 [
50 "8",
51 "seed"
52 ],
53 [
54 "8",
55 "control_after_generate"
56 ],
57 [
58 "4",
59 "unet_name"
60 ],
61 [
62 "6",
63 "vae_name"
64 ],
65 [
66 "1",
67 "clip_name1"
68 ],
69 [
70 "1",
71 "clip_name2"
72 ]
73 ]
74 },
75 "widgets_values": []
76 },
77 {
78 "id": 11,
79 "type": "SaveVideo",
80 "pos": [
81 1340,
82 730
83 ],
84 "size": [
85 770,
86 640
87 ],
88 "flags": [],
89 "order": 3,
90 "mode": 0,
91 "inputs": [
92 {
93 "name": "video",
94 "type": "VIDEO",
95 "link": 15
96 }
97 ],
98 "outputs": [],
99 "properties": {
100 "Node name for S&R": "SaveVideo",
101 "cnr_id": "comfy-core",
102 "ver": "0.3.68"
103 },
104 "widgets_values": [
105 "video/Kandinsky5",
106 "mp4",
107 "h264"
108 ]
109 },
110 {
111 "id": 14,
112 "type": "MarkdownNote",
113 "pos": [
114 620,
115 1290
116 ],
117 "size": [
118 240,
119 90
120 ],
121 "flags": {
122 "collapsed": false
123 },
124 "order": 1,
125 "mode": 0,
126 "inputs": [],
127 "outputs": [],
128 "title": "Note: Video Size",
129 "properties": [],
130 "widgets_values": [
131 "768x512x121 for 5s or 768x512x241 for 10s, Width and Height should be divisible by 128 for 10s model. In case of image 2 video aspect ratio of result is close to start image ratio and area is close to 768*512"
132 ],
133 "color": "#222",
134 "bgcolor": "#000"
135 },
136 {
137 "id": 15,
138 "type": "MarkdownNote",
139 "pos": [
140 400,
141 740
142 ],
143 "size": [
144 460,
145 480
146 ],
147 "flags": [],
148 "order": 2,
149 "mode": 0,
150 "inputs": [],
151 "outputs": [],
152 "title": "For Local User",
153 "properties": [],
154 "widgets_values": [
155 "\n## Model links\n\n- [Hugging Face:Comfy-Org/HunyuanVideo_1.5_repackaged](https://huggingface.co/Comfy-Org/HunyuanVideo_1.5_repackaged)\n\n**ModelScope mirrors** (alternative repo location)\n\n- [ModelScope:Comfy-Org/HunyuanVideo_1.5_repackaged](https://modelscope.cn/models/Comfy-Org/HunyuanVideo_1.5_repackaged)\n\nYou can find the step distilled model weight here: [kandinsky-50-video-lite](https://huggingface.co/collections/kandinskylab/kandinsky-50-video-lite)\n\n**text_encoders**\n\n- [qwen_2.5_vl_7b_fp8_scaled.safetensors](https://huggingface.co/Comfy-Org/HunyuanVideo_1.5_repackaged/resolve/main/split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors) (8.74 GB)\n- [clip_l.safetensors](https://huggingface.co/comfyanonymous/flux_text_encoders/resolve/main/clip_l.safetensors?download=true) (234.7 MB)\n\n**diffusion_models**\n\n- [kandinsky5lite_t2v_sft_5s.safetensors](https://huggingface.co/kandinskylab/Kandinsky-5.0-T2V-Lite-sft-5s/resolve/main/model/kandinsky5lite_t2v_sft_5s.safetensors) (4.26 GB)\n\n**vae**\n\n- [hunyuan_video_vae_bf16.safetensors](https://huggingface.co/Kijai/HunyuanVideo_comfy/resolve/main/hunyuan_video_vae_bf16.safetensors) (470.1 MB)\n\nModel Storage Location\n\n```\n📂 ComfyUI/\n├── 📂 models/\n│ ├── 📂 text_encoders/\n│ │ ├── qwen_2.5_vl_7b_fp8_scaled.safetensors\n│ │ └── clip_l.safetensors\n│ ├── 📂 diffusion_models/\n│ │ └── kandinsky5lite_t2v_sft_5s.safetensors\n│ └── 📂 vae/\n│ └── hunyuan_video_vae_bf16.safetensors\n```\n\n## Report issue\n\nIf you found any issues when running this workflow, [report template issue here](https://github.com/Comfy-Org/workflow_templates/issues)\n"
156 ],
157 "color": "#222",
158 "bgcolor": "#000"
159 }
160 ],
161 "links": [
162 [
163 15,
164 12,
165 0,
166 11,
167 0,
168 "VIDEO"
169 ]
170 ],
171 "groups": [],
172 "definitions": {
173 "subgraphs": [
174 {
175 "id": "7aad998c-49e7-433f-bfb9-b1ac2680aa9e",
176 "version": 1,
177 "state": {
178 "lastGroupId": 2,
179 "lastNodeId": 11,
180 "lastLinkId": 14,
181 "lastRerouteId": 0
182 },
183 "revision": 0,
184 "config": [],
185 "name": "Kandinsky5 Text to Video",
186 "inputNode": {
187 "id": -10,
188 "bounding": [
189 150,
190 855.6086185239999,
191 120,
192 40
193 ]
194 },
195 "outputNode": {
196 "id": -20,
197 "bounding": [
198 2543.3334996748745,
199 845.6086185239999,
200 120,
201 60
202 ]
203 },
204 "inputs": [],
205 "outputs": [
206 {
207 "id": "3269ff3b-db7a-40e4-b5b2-52ad249ad50b",
208 "name": "VIDEO",
209 "type": "VIDEO",
210 "linkIds": [
211 14
212 ],
213 "localized_name": "VIDEO",
214 "pos": [
215 20,
216 20
217 ]
218 }
219 ],
220 "widgets": [],
221 "nodes": [
222 {
223 "id": 2,
224 "type": "CLIPTextEncode",
225 "pos": [
226 884.6464438178214,
227 917.8318920966483
228 ],
229 "size": [
230 479.97395833333337,
231 210.91145833333334
232 ],
233 "flags": [],
234 "order": 3,
235 "mode": 0,
236 "inputs": [
237 {
238 "localized_name": "clip",
239 "name": "clip",
240 "type": "CLIP",
241 "link": 1
242 }
243 ],
244 "outputs": [
245 {
246 "localized_name": "CONDITIONING",
247 "name": "CONDITIONING",
248 "type": "CONDITIONING",
249 "links": [
250 4
251 ]
252 }
253 ],
254 "title": "CLIP Text Encode (Negative Prompt)",
255 "properties": {
256 "Node name for S&R": "CLIPTextEncode"
257 },
258 "widgets_values": [
259 ""
260 ],
261 "color": "#223",
262 "bgcolor": "#335"
263 },
264 {
265 "id": 1,
266 "type": "DualCLIPLoader",
267 "pos": [
268 344.6467588070548,
269 957.8319160455486
270 ],
271 "size": [
272 500,
273 130
274 ],
275 "flags": [],
276 "order": 0,
277 "mode": 0,
278 "inputs": [],
279 "outputs": [
280 {
281 "localized_name": "CLIP",
282 "name": "CLIP",
283 "type": "CLIP",
284 "links": [
285 1,
286 6
287 ]
288 }
289 ],
290 "properties": {
291 "Node name for S&R": "DualCLIPLoader",
292 "models": [
293 {
294 "name": "qwen_2.5_vl_7b_fp8_scaled.safetensors",
295 "url": "https://huggingface.co/Comfy-Org/HunyuanVideo_1.5_repackaged/resolve/main/split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors",
296 "directory": "text_encoders"
297 },
298 {
299 "name": "clip_l.safetensors",
300 "url": "https://huggingface.co/comfyanonymous/flux_text_encoders/resolve/main/clip_l.safetensors?download=true",
301 "directory": "text_encoders"
302 }
303 ]
304 },
305 "widgets_values": [
306 "qwen_2.5_vl_7b_fp8_scaled.safetensors",
307 "clip_l.safetensors",
308 "kandinsky5",
309 "default"
310 ]
311 },
312 {
313 "id": 6,
314 "type": "VAELoader",
315 "pos": [
316 344.6467588070548,
317 797.8318557110248
318 ],
319 "size": [
320 490,
321 58
322 ],
323 "flags": [],
324 "order": 1,
325 "mode": 0,
326 "inputs": [],
327 "outputs": [
328 {
329 "localized_name": "VAE",
330 "name": "VAE",
331 "type": "VAE",
332 "links": [
333 5,
334 12
335 ]
336 }
337 ],
338 "properties": {
339 "Node name for S&R": "VAELoader",
340 "models": [
341 {
342 "name": "hunyuan_video_vae_bf16.safetensors",
343 "url": "https://huggingface.co/Kijai/HunyuanVideo_comfy/resolve/main/hunyuan_video_vae_bf16.safetensors",
344 "directory": "vae"
345 }
346 ]
347 },
348 "widgets_values": [
349 "hunyuan_video_vae_bf16.safetensors"
350 ]
351 },
352 {
353 "id": 7,
354 "type": "CLIPTextEncode",
355 "pos": [
356 884.6464438178214,
357 607.831812875903
358 ],
359 "size": [
360 479.97395833333337,
361 210.91145833333334
362 ],
363 "flags": [],
364 "order": 4,
365 "mode": 0,
366 "inputs": [
367 {
368 "localized_name": "clip",
369 "name": "clip",
370 "type": "CLIP",
371 "link": 6
372 }
373 ],
374 "outputs": [
375 {
376 "localized_name": "CONDITIONING",
377 "name": "CONDITIONING",
378 "type": "CONDITIONING",
379 "links": [
380 3
381 ]
382 }
383 ],
384 "title": "CLIP Text Encode (Positive Prompt)",
385 "properties": {
386 "Node name for S&R": "CLIPTextEncode"
387 },
388 "widgets_values": [
389 "Rim light, side light, soft light, medium close-up, dusk, sunset, central composition, warm tones with low saturation, telephoto lens. A woman with fluffy brown curly hair stands elegantly in front of a magnificent stained glass window. She is wearing a flowing white dress, her hair neatly combed back, and her soft facial contours are gently illuminated by the colorful light filtering through the window from outside. The woman is talking to someone off-camera, yet there is a hint of sadness in her eyes, adding a layer of depth to her mysterious temperament. The background is dim, with a strong contrast between light and shadow, further emphasizing the tension of the character's emotions. The stained glass, under the glow of the setting sun, casts colorful light and shadows, enhancing the artistic sense and atmosphere of the overall picture."
390 ],
391 "color": "#232",
392 "bgcolor": "#353"
393 },
394 {
395 "id": 9,
396 "type": "VAEDecode",
397 "pos": [
398 2173.3334719192007,
399 620.0000154949549
400 ],
401 "size": [
402 230,
403 46
404 ],
405 "flags": [],
406 "order": 8,
407 "mode": 0,
408 "inputs": [
409 {
410 "localized_name": "samples",
411 "name": "samples",
412 "type": "LATENT",
413 "link": 11
414 },
415 {
416 "localized_name": "vae",
417 "name": "vae",
418 "type": "VAE",
419 "link": 12
420 }
421 ],
422 "outputs": [
423 {
424 "localized_name": "IMAGE",
425 "name": "IMAGE",
426 "type": "IMAGE",
427 "links": [
428 13
429 ]
430 }
431 ],
432 "properties": {
433 "Node name for S&R": "VAEDecode"
434 },
435 "widgets_values": []
436 },
437 {
438 "id": 3,
439 "type": "ModelSamplingSD3",
440 "pos": [
441 1434.6466135403523,
442 597.8318778108319
443 ],
444 "size": [
445 320,
446 58
447 ],
448 "flags": [],
449 "order": 5,
450 "mode": 0,
451 "inputs": [
452 {
453 "localized_name": "model",
454 "name": "model",
455 "type": "MODEL",
456 "link": 2
457 }
458 ],
459 "outputs": [
460 {
461 "localized_name": "MODEL",
462 "name": "MODEL",
463 "type": "MODEL",
464 "slot_index": 0,
465 "links": [
466 7
467 ]
468 }
469 ],
470 "properties": {
471 "Node name for S&R": "ModelSamplingSD3",
472 "cnr_id": "comfy-core",
473 "ver": "0.3.68"
474 },
475 "widgets_values": [
476 5
477 ]
478 },
479 {
480 "id": 8,
481 "type": "KSampler",
482 "pos": [
483 1814.6469961959506,
484 617.8318203813332
485 ],
486 "size": [
487 335.9375,
488 603.3854166666667
489 ],
490 "flags": [],
491 "order": 7,
492 "mode": 0,
493 "inputs": [
494 {
495 "localized_name": "model",
496 "name": "model",
497 "type": "MODEL",
498 "link": 7
499 },
500 {
501 "localized_name": "positive",
502 "name": "positive",
503 "type": "CONDITIONING",
504 "link": 8
505 },
506 {
507 "localized_name": "negative",
508 "name": "negative",
509 "type": "CONDITIONING",
510 "link": 9
511 },
512 {
513 "localized_name": "latent_image",
514 "name": "latent_image",
515 "type": "LATENT",
516 "link": 10
517 }
518 ],
519 "outputs": [
520 {
521 "localized_name": "LATENT",
522 "name": "LATENT",
523 "type": "LATENT",
524 "links": [
525 11
526 ]
527 }
528 ],
529 "properties": {
530 "Node name for S&R": "KSampler",
531 "cnr_id": "comfy-core",
532 "ver": "0.3.71"
533 },
534 "widgets_values": [
535 102704842425568,
536 "randomize",
537 50,
538 5,
539 "euler_ancestral",
540 "beta",
541 1
542 ]
543 },
544 {
545 "id": 5,
546 "type": "Kandinsky5ImageToVideo",
547 "pos": [
548 1434.6466135403523,
549 757.8319026842786
550 ],
551 "size": [
552 330,
553 190
554 ],
555 "flags": [],
556 "order": 6,
557 "mode": 0,
558 "inputs": [
559 {
560 "localized_name": "positive",
561 "name": "positive",
562 "type": "CONDITIONING",
563 "link": 3
564 },
565 {
566 "localized_name": "negative",
567 "name": "negative",
568 "type": "CONDITIONING",
569 "link": 4
570 },
571 {
572 "localized_name": "vae",
573 "name": "vae",
574 "type": "VAE",
575 "link": 5
576 },
577 {
578 "localized_name": "start_image",
579 "name": "start_image",
580 "shape": 7,
581 "type": "IMAGE",
582 "link": null
583 }
584 ],
585 "outputs": [
586 {
587 "localized_name": "positive",
588 "name": "positive",
589 "type": "CONDITIONING",
590 "links": [
591 8
592 ]
593 },
594 {
595 "localized_name": "negative",
596 "name": "negative",
597 "type": "CONDITIONING",
598 "links": [
599 9
600 ]
601 },
602 {
603 "localized_name": "latent",
604 "name": "latent",
605 "type": "LATENT",
606 "links": [
607 10
608 ]
609 },
610 {
611 "localized_name": "cond_latent",
612 "name": "cond_latent",
613 "type": "LATENT",
614 "links": []
615 }
616 ],
617 "properties": {
618 "Node name for S&R": "Kandinsky5ImageToVideo"
619 },
620 "widgets_values": [
621 768,
622 512,
623 121,
624 1
625 ]
626 },
627 {
628 "id": 10,
629 "type": "CreateVideo",
630 "pos": [
631 2173.3334996748745,
632 770.00011240937
633 ],
634 "size": [
635 310,
636 78
637 ],
638 "flags": {
639 "collapsed": false
640 },
641 "order": 9,
642 "mode": 0,
643 "inputs": [
644 {
645 "localized_name": "images",
646 "name": "images",
647 "type": "IMAGE",
648 "link": 13
649 },
650 {
651 "localized_name": "audio",
652 "name": "audio",
653 "shape": 7,
654 "type": "AUDIO",
655 "link": null
656 }
657 ],
658 "outputs": [
659 {
660 "localized_name": "VIDEO",
661 "name": "VIDEO",
662 "type": "VIDEO",
663 "links": [
664 14
665 ]
666 }
667 ],
668 "properties": {
669 "Node name for S&R": "CreateVideo",
670 "cnr_id": "comfy-core",
671 "ver": "0.3.68"
672 },
673 "widgets_values": [
674 24
675 ]
676 },
677 {
678 "id": 4,
679 "type": "UNETLoader",
680 "pos": [
681 344.6467588070548,
682 607.831812875903
683 ],
684 "size": [
685 490,
686 90
687 ],
688 "flags": [],
689 "order": 2,
690 "mode": 0,
691 "inputs": [],
692 "outputs": [
693 {
694 "localized_name": "MODEL",
695 "name": "MODEL",
696 "type": "MODEL",
697 "slot_index": 0,
698 "links": [
699 2
700 ]
701 }
702 ],
703 "properties": {
704 "Node name for S&R": "UNETLoader",
705 "cnr_id": "comfy-core",
706 "ver": "0.3.68",
707 "models": [
708 {
709 "name": "kandinsky5lite_t2v_sft_5s.safetensors",
710 "url": "https://huggingface.co/kandinskylab/Kandinsky-5.0-T2V-Lite-sft-5s/resolve/main/model/kandinsky5lite_t2v_sft_5s.safetensors",
711 "directory": "diffusion_models"
712 }
713 ]
714 },
715 "widgets_values": [
716 "kandinsky5lite_t2v_sft_5s.safetensors",
717 "default"
718 ]
719 }
720 ],
721 "groups": [
722 {
723 "id": 1,
724 "title": "Load models",
725 "bounding": [
726 330,
727 530,
728 514.6034734835969,
729 604.32
730 ],
731 "color": "#3f789e",
732 "font_size": 24,
733 "flags": []
734 },
735 {
736 "id": 2,
737 "title": "Prompt",
738 "bounding": [
739 870,
740 530,
741 500,
742 608.3333333333334
743 ],
744 "color": "#3f789e",
745 "font_size": 24,
746 "flags": []
747 }
748 ],
749 "links": [
750 {
751 "id": 1,
752 "origin_id": 1,
753 "origin_slot": 0,
754 "target_id": 2,
755 "target_slot": 0,
756 "type": "CLIP"
757 },
758 {
759 "id": 6,
760 "origin_id": 1,
761 "origin_slot": 0,
762 "target_id": 7,
763 "target_slot": 0,
764 "type": "CLIP"
765 },
766 {
767 "id": 11,
768 "origin_id": 8,
769 "origin_slot": 0,
770 "target_id": 9,
771 "target_slot": 0,
772 "type": "LATENT"
773 },
774 {
775 "id": 12,
776 "origin_id": 6,
777 "origin_slot": 0,
778 "target_id": 9,
779 "target_slot": 1,
780 "type": "VAE"
781 },
782 {
783 "id": 2,
784 "origin_id": 4,
785 "origin_slot": 0,
786 "target_id": 3,
787 "target_slot": 0,
788 "type": "MODEL"
789 },
790 {
791 "id": 7,
792 "origin_id": 3,
793 "origin_slot": 0,
794 "target_id": 8,
795 "target_slot": 0,
796 "type": "MODEL"
797 },
798 {
799 "id": 8,
800 "origin_id": 5,
801 "origin_slot": 0,
802 "target_id": 8,
803 "target_slot": 1,
804 "type": "CONDITIONING"
805 },
806 {
807 "id": 9,
808 "origin_id": 5,
809 "origin_slot": 1,
810 "target_id": 8,
811 "target_slot": 2,
812 "type": "CONDITIONING"
813 },
814 {
815 "id": 10,
816 "origin_id": 5,
817 "origin_slot": 2,
818 "target_id": 8,
819 "target_slot": 3,
820 "type": "LATENT"
821 },
822 {
823 "id": 3,
824 "origin_id": 7,
825 "origin_slot": 0,
826 "target_id": 5,
827 "target_slot": 0,
828 "type": "CONDITIONING"
829 },
830 {
831 "id": 4,
832 "origin_id": 2,
833 "origin_slot": 0,
834 "target_id": 5,
835 "target_slot": 1,
836 "type": "CONDITIONING"
837 },
838 {
839 "id": 5,
840 "origin_id": 6,
841 "origin_slot": 0,
842 "target_id": 5,
843 "target_slot": 2,
844 "type": "VAE"
845 },
846 {
847 "id": 13,
848 "origin_id": 9,
849 "origin_slot": 0,
850 "target_id": 10,
851 "target_slot": 0,
852 "type": "IMAGE"
853 },
854 {
855 "id": 14,
856 "origin_id": 10,
857 "origin_slot": 0,
858 "target_id": -20,
859 "target_slot": 0,
860 "type": "VIDEO"
861 }
862 ],
863 "extra": {
864 "workflowRendererVersion": "LG"
865 }
866 }
867 ]
868 },
869 "config": [],
870 "extra": {
871 "ds": {
872 "scale": 0.7271498427385472,
873 "offset": [
874 9.043333262936898,
875 -500.76099538866106
876 ]
877 },
878 "frontendVersion": "1.38.6",
879 "workflowRendererVersion": "LG"
880 },
881 "version": 0.4
882}